Perre v Apand Pty Ltd

GLEESON CJ. The issue to be decided is whether the respondent, whose careless conduct resulted in financial loss to others, unconnected with physical injury to their persons or property, owed them a duty of care such as to sustain an action for damages for negligence.

The facts of the case are set out in the reasons for judgment of other members of the Court. In broadest outline, they are as follows. In a rural locality in South Australia, a number of farmers grew potatoes, some for export to Western Australia. The respondent negligently introduced a form of disease, known as bacterial wilt, onto the land of one farmer. The Western Australian regulations imposed a prohibition on the importation into Western Australia, not only of potatoes grown on land known to be affected by the disease, but also of potatoes grown on land within a certain distance of affected land. The appellants were involved, in various ways, in potato growing on such land, and claimed to suffer financial loss. The issue is whether the respondent, whose conduct is, for the purposes of argument, assumed to have caused harm to each of the appellants, owed a duty of care to all or any of them. In the Federal Court at first instance , and in the Full Court , that question was answered in the negative as to all appellants.

Although the appellants, considered collectively, owned, or had other interests in, land which was in the neighbourhood of land directly affected by the respondent's negligent conduct, no property belonging to, or used by, the appellants suffered any physical harm. The loss allegedly suffered by each appellant was of a kind conventionally described as pure economic loss , although the utility of considering such loss without further categorisation has been questioned by some judges and commentators .

In Caltex Oil (Australia) Pty Ltd v The Dredge "Willemstad" all the members of the Court, except Murphy J, accepted that there is no general rule that one person owes to another a duty to take care not to cause reasonably foreseeable financial harm . The consequences of such a rule would be intolerable. However, as the decision in that case showed, and as had previously been shown in Hedley Byrne & Co Ltd v Heller & Partners Ltd , there are circumstances in which the law recognises a duty of care such as will permit recovery of pure economic loss.

There are at least three considerations which have been, and will remain, influential in restraining acceptance of such a duty of care in particular cases, or categories of case. First, bearing in mind the expansive application which has been given to the concept of reasonable foreseeability in relation to physical injury to person or property, a duty to avoid any reasonably foreseeable financial harm needs to be constrained by "some intelligible limits to keep the law of negligence within the bounds of common sense and practicality" . Secondly, to permit recovery of foreseeable economic loss, which may or may not occur in a commercial setting, for any negligent conduct, may interfere with freedoms, controls and limitations established both by common law and statute in many legal contexts. Thirdly, in those cases where the loss occurs in a commercial setting, a third party, C, may suffer financial harm as a result of conduct which is regulated by a contract between A and B. It may be that the consequences of such conduct, as between A and B, are governed and limited by the contract. This is a problem which commonly occurs in relation to maritime claims, and may help to explain the strictness with which an exclusionary rule has been applied in shipping cases .

Another matter of concern is the lack of precision in the concept of financial or economic loss. Physical injury to person or property is usually readily identifiable, even if it may take time to manifest itself. However, the concept of financial or economic loss or harm is wide enough to comprehend a variety of circumstances or contingencies, some of which may be indirect and difficult to identify or measure. Suppose, for example, that a child's parents are killed as a result of the negligent conduct of another. In many jurisdictions there are statutory provisions which govern entitlement to compensation . However, if the matter were at large, how would a court set about identifying, and estimating, the kinds of financial loss which might sound in damages? What kinds of detriment, harm, or disadvantage, would be treated as "financial loss"? The law of tort is a blunt instrument for providing a remedy for many kinds of harm which may be suffered as a consequence of someone else's carelessness, and which are capable of being described as financial.

If there once was a bright line rule which absolutely prevented recognition of a duty of care in any case where the negligent conduct of one person caused financial loss to another, not associated with injury to the other's person or property, and which assigned claims to recover such loss to the field of contract rather than tort, the line gave way in an area where there is a clear potential for carelessness to cause financial harm: negligent misstatements made to a person who, to the knowledge of the maker of the statement, relies upon the advice or information provided. However, there is no convincing reason why conveying advice or information should be treated as the solitary exception to an otherwise absolute exclusionary rule.

Once the exclusionary rule ceased to be a bright line rule, it lost one of its principal justifications. Nevertheless, the considerations underlying the rule remain cogent, even if they are no longer seen as absolutely compelling. Courts have found difficulty in proposing an alternative general rule which makes better sense and which, at the same time, pays due regard to the problems earlier mentioned.

The solution does not lie in what is sometimes described as the three-stage "test" said to have been formulated by Lord Bridge of Harwich in Caparo Industries Plc v Dickman . Lord Bridge never said it did. He said it did not. In the much quoted passage in his Lordship's speech where he referred to the necessary ingredients of foreseeability, proximity, and a situation in which the court considers it fair, just and reasonable that the law should impose a duty, he immediately went on to say that "the concepts of proximity and fairness ... are not susceptible of any such precise definition as would be necessary to give them utility as practical tests, but amount in effect to little more than convenient labels to attach to the features of different specific situations which, on a detailed examination of all the circumstances, the law recognises pragmatically as giving rise to a duty of care of a given scope". He also quoted with approval an observation of Brennan J concerning incremental development of the law. In the same case, Lord Oliver of Aylmerton , whose speech has been equally influential in later cases, said that "to search for any single formula which will serve as a general test of liability is to pursue a will-o'-the wisp."

In Caparo , Lord Oliver emphasised that, in this field of discourse, the mere foreseeability of possible damage, without some further control, (which he summarised as "proximity", after explaining what he meant by that term), would not be useful as the test of liability . At the same time, however, his Lordship made it clear that "in some cases the degree of foreseeability is such that it is from that alone that the requisite proximity can be deduced." In relation to the giving of advice or information, questions of reliance and actual foresight of the possibility of harm, (or, what is the same thing, the foresight that a reasonable person would have), are closely related. Moreover, knowledge (actual, or that which a reasonable person would have) of an individual, or an ascertainable class of persons, who is or are reliant, and therefore vulnerable, is a significant factor in establishing a duty of care.

Vulnerability can arise from circumstances other than reliance. In Caltex, the obvious vulnerability of a specific plaintiff was influential in a number of the judgments. This was not merely an arbitrary method of solving the problem of potentially indeterminate liability. It was an application of what Lord Oliver later discussed as the idea that in a given case, the degree (and nature) of foreseeability may have an important bearing on whether there is a duty of care.

I agree with the reasons given by Gummow J for concluding that, in the present case, the respondent owed the appellants a duty of care. In particular, I would emphasise the following matters.

The acknowledgment, in the internal communications of the respondent, that there was a need to be careful so as not to damage the interests of those involved in potato growing on land within 20 km of a farm that might be affected by bacterial wilt, is not merely a matter of legally irrelevant prejudice. It shows actual foresight of the likelihood of harm, and knowledge of an ascertainable class of vulnerable persons.

If it had been the fault of the Sparnons, rather than the respondent, that there was an outbreak of bacterial wilt on their land, then the resulting interference with the use and enjoyment of neighbouring land would not have gone without remedy. The respondent controlled the activity on the Sparnons' land, which resulted in land and facilities within 20 km being treated by Western Australian law, for practical purposes, as if contaminated.

The reasoning in the Federal Court was strongly influenced by the consideration that there was seen to be no rational basis for distinguishing between the position of the present appellants and that of an indeterminate number of other people. That was unquestionably an important issue to address. A decision that a duty of care is owed to some who suffer financial harm as a consequence of negligent conduct, but not to others, is only just if it is capable of being explained on a rational basis. As the judgments of other members of this Court show, to an extent the Federal Court's conclusion was based upon a view of the facts, and in particular of the critical aspect of the negligent conduct, which cast the net of potential liability too wide. Furthermore, the combination of circumstances involving the use and ownership or enjoyment of land, the physical propinquity of such land to the Sparnons' land, the known vulnerability of people in the position of the appellants, and the control exercised by the respondent over the relevant activity on the Sparnons' land, is unlikely to apply to an extent sufficient to warrant an apprehension of indeterminate liability.

I agree that the appeal should be allowed, and orders made as proposed by Gaudron J.

GAUDRON J. The relevant facts and the history of these proceedings are set out in other judgments. I shall repeat them only to the extent necessary to make clear my reasons for concluding that the appeal in this matter should be allowed. The relationship between the appellants

In order to identify the interests of the appellants which were infringed by the respondent's transmission of seed potatoes affected by bacterial wilt to South Australia for planting on the property of Mr and Mrs David Sparnon and their son, Michael ("the Sparnon property"), it is necessary to say something of the relationship which is said to have existed between the various appellants when bacterial wilt was detected on that property in April 1992. Until then, none of the appellants suffered any damage; and prima facie, at least, if damage was suffered, it was suffered when the disease was detected, for it was detection of the disease that prevented the appellants from selling potatoes into the Western Australian market or using their land and equipment to grow and pack potatoes for that market.

What follows with respect to the relationship between the appellants is taken from the Ninth Statement of Claim which, unfortunately, does not seem to have been drafted with a view either to elucidating the precise nature of the appellants' legal arrangements or to particularising the damage that each claims to have suffered. However, the relationships appear to centre on two corporate appellants, Warruga Farms Pty Ltd ("Warruga Farms") and Perre's Vineyards Pty Ltd ("Perre's Vineyards"), and a partnership known as the Rangara joint venture.

Warruga Farms grew potatoes on land three kilometres distant from the Sparnon property using, it seems, its own machinery and equipment. The Crown leasehold in the land was held by the first to eighth appellants, they being the fourth to eleventh applicants in the action. Although it is not clear from the pleadings, it seems likely that Warruga Farms owned the potatoes grown on the land. It is convenient to proceed on the assumption that that was so. Additionally and as part of its business, Warruga Farms acquired potatoes grown by the Rangara joint venture.

Warruga Farms washed and packed the potatoes it grew and, apparently, those it acquired from the Rangara joint venture in a purpose built packing shed on land leased by Perre's Vineyards three kilometres distant from the Sparnon property. Warruga Farms used that packing shed pursuant to a tenancy at will for which it paid between $16,000 and $20,000 per year. Perre's Vineyards held a Crown lease over the land on which the packing shed was located.

The Rangara joint venture is a partnership between Pasquale and Grace Perre, the twelfth and thirteenth appellants and Rangara Pty Limited ("Rangara"), a company in which the shares are held by Francesco and Maria Perre, the fourteenth and fifteenth appellants. Although it is not entirely clear, it seems that, in its potato growing operations, the Rangara joint venture used equipment which Rangara owned and, also, some owned by Francesco and Maria Perre. The Rangara joint venture grew potatoes on two blocks of land which were each two kilometres distant from the Sparnon property and which were owned, respectively, by Pasquale and Grace Perre and by Francesco and Maria Perre.

Again, it seems likely that the Rangara joint venture owned the potatoes it grew until they were sold to Warruga Farms although the pleadings do not disclose when, if at all, Warruga Farms obtained the property in them. It is convenient to proceed on the assumption that some potatoes were owned by the Rangara joint venture when bacterial wilt was discovered on the Sparnon property. Liability for economic loss

As the other judgments make clear, bacterial wilt neither infected the potatoes grown by Warruga Farms nor those grown by the Rangara joint venture. Nor did it spread to any of the lands on which the potatoes were grown or packed. The case is, thus, a case of pure economic loss, in the sense that the loss in issue does not result from any physical injury. Rather, the loss occurred because, by reason of the outbreak of bacterial wilt on the Sparnon property, Warruga Farms and the Rangara joint venture were unable to sell their potatoes into the Western Australian market and, for a period of five years, the other appellants were unable to use their land or equipment for the production or preparation of potatoes for that market.

The law as to liability for economic loss is a "comparatively new and developing area of the law of negligence." It has not yet developed to a stage where there has been enunciated a governing principle applicable in all cases . Perhaps it never will. Not surprisingly, given the present stage of development, different approaches have been advanced as to the way in which claims for which there is no legal precedent should be dealt with.

The first of the various approaches propounded in this area of the law was formulated by Lord Wilberforce in Anns v Merton London Borough Council . That approach involved two considerations: firstly, foreseeability and, secondly, considerations which "negative, or ... reduce or limit the scope of the duty or the class of person to whom it is owed or the damages to which a breach of it may give rise" . In Sutherland Shire Council v Heyman , Brennan J favoured the development of "novel categories of negligence incrementally and by analogy with established categories" , an approach later adopted by the House of Lords in Murphy v Brentwood District Council . And in Pyrenees Shire Council v Day , Kirby J advocated the three stage test which is now generally applied in England . That test involves firstly, foreseeability, secondly, the existence of a relationship between the parties of "proximity" or "neighbourhood" and finally, a consideration of policy to determine whether it is "fair, just and reasonable" to impose the duty of care in question .

It is clear that foreseeability does not, of itself, suffice to render a defendant liable for negligently inflicted economic loss. This notwithstanding, the notion of proximity, which has generally been adopted by this Court to describe the special feature or features that attract a duty of care in economic loss cases , has been criticised as being incapable of constituting a universal criterion of liability and, also, as having only limited utility in determining whether there exists a duty of care in a particular case . It may well be that, at this stage, the notion of proximity can serve no purpose beyond signifying that it is necessary to identify a factor or factors of special significance in addition to the foreseeability of harm before the law will impose liability for the negligent infliction of economic loss . Categories of case: negligent misstatement

It seems to me that much judicial effort has been devoted to, but little gained by, criticising the approaches that have been advanced or theories propounded in this area of the law. Rather, it seems likely that, in time, there will develop a sufficient body of case law from which it is possible to discern different categories for which the special circumstances that call a duty of care into existence can be articulated. That is what Stephen J predicted in Caltex Oil (Australia) Pty Ltd v The Dredge "Willemstad" .

So far as concerns the category of negligent misstatement - more accurately, the failure to provide information or advice or, usually, the failure to provide accurate information or advice - the prediction of Stephen J in Caltex Oil has largely been borne out. That category was impliedly recognised as a discrete category in San Sebastian Pty Ltd v The Minister , and expressly recognised as such by Brennan CJ and, also, by Dawson J in Hill v Van Erp .

So far as concerns negligent misstatement, the circumstances which attract a duty of care are "known reliance (or dependence) or the assumption of responsibility or a combination of the two" , the word "known" including circumstances in which reliance or dependence ought to be known . And in Esanda Finance Corporation Ltd v Peat Marwick Hungerfords , it was not pleaded that the auditors in question knew or ought to have known that a finance provider would rely on their audited statement of accounts and, thus, it was held, on the pleadings, that no duty of care was owed by the auditors to the finance provider . Policy considerations

It seems to me possible to discern another category for which the circumstances that attract a duty of care can now be articulated. Although that category does not cover this precise case, it is one by analogy with which this case is, in my view, to be decided. Before turning to that second category, however, it is convenient to note two policy considerations that are frequently invoked in this area of the law.

The first policy consideration is "the law's concern to avoid the imposition of liability 'in an indeterminate amount for an indeterminate time to an indeterminate class'." It is important to remember that that is a policy consideration, not a rule of law. Thus, it is not necessarily fatal to the recognition of a duty of care that the duty is owed to a class whose members cannot be identified with complete accuracy.

The second policy consideration is that, in a competitive commercial environment, "a duty to take reasonable care to avoid causing mere economic loss to another ... may be inconsistent with community standards in relation to what is ordinarily legitimate in the pursuit of personal advantage." It is because of this that the law requires some special factor or factors before it will impose a duty of care in protection of commercial interests, opportunities or, even, advantages. However, the factor or factors which will attract liability may be of a somewhat broader character when economic loss results from the destruction or impairment of a legal right. Categories of case: protection of legal rights

As I pointed out in Hawkins v Clayton , "[t]he law of tort already protects contractual rights from intentional interference ... [and the] torts of trespass, conversion, detinue and slander of title are intimately concerned with the protection of legal rights" . And that is so even where loss or impairment results in pure economic loss. In that context, it is not surprising that, in some situations, the law of negligence may be invoked where economic loss is suffered in consequence of the loss or impairment of a legal right. On the contrary, there are situations (for example, the retained solicitor whose negligence results in a cause of action becoming statute barred) in which it would be surprising if the law of negligence could not be invoked.

It is in the area of legal rights that, in my view, there is to be discerned a second discrete category of liability for pure economic loss. Indeed, in some situations, the existence of a duty of care with respect to another's legal rights may be so obvious as to pass without argument. Thus, for example, there was no issue in Bennett v Minister of Community Welfare but that "the circumstances of the guardianship and the injury to the appellant while under the care and control of the [servant or agent of the Minister] gave rise to a common law duty on the part of the latter to take reasonable care to ensure that the appellant did not suffer economic loss by not being advised of his rights in respect of that injury."

The legal right involved in Bennett was the right to bring an action for damages for personal injury. The appellant, who was, in effect, a ward of State, was dependent on his guardian for information and advice as to his rights in that regard. By corollary, his guardian was in a position to control the appellant's exercise of his right to seek damages. The circumstances involved in Hawkins and in Hill are comparable. In Hawkins , the legal right involved was that of the named executor to obtain probate of a will. The solicitor, who had custody of the will, was held to have been under a duty of care to take reasonable steps to inform the named executor of its terms. In Hill , the right involved was that of a beneficiary to receive a testamentary gift intended by a testatrix, the right having been defeated by a solicitor's failure to ensure that the will was not witnessed by the beneficiary's husband.

In Hill , both Gummow J and I saw the solicitor's control over the realisation of the intentions of the testatrix as a special factor warranting the imposition of a duty of care . And Dawson J, with whom Toohey J agreed, placed emphasis on the fact that "the intended beneficiary's interests [were] totally and unavoidably dependent upon the proper performance of a function within the sole province of the solicitor" , which, of course, is simply the corollary of the solicitor's control. And as I said in Hill , it seems to me that Hawkins is also to be explained in terms of the solicitor's control, as a matter of practical reality, over the exercise by the executor of his right to apply for probate.

Where a person is in a position to control the exercise or enjoyment by another of a legal right, that position of control and, by corollary, the other's dependence on the person with control are, in my view, special factors or, which is the same thing, give rise to a special relationship of "proximity" or "neighbourhood" such that the law will impose liability upon the person with control if his or her negligent act or omission results in the loss or impairment of that right and is, thereby, productive of economic loss. The appeal

As earlier indicated, the circumstances of this case are analogous but not truly comparable with those in Bennett , Hawkins and Hill . The main difference is that, in those cases, the right involved was a right peculiar to the plaintiff. In this case, the rights involved are general rights, namely, the right to sell potatoes in the Western Australian market and the right to use one's land and equipment for the production of potatoes for that purpose. So far as concerns persons not resident in Western Australia, those rights can only be exercised or enjoyed by those who satisfy the conditions imposed by the law of Western Australia for the entry of potatoes into that State. More precisely, the rights are rights which can be exercised or enjoyed only by members of a particular class. And in fact, they were exercised or enjoyed by some only of that class.

The consideration that the legal rights involved in this case are rights which attach to members of a class rather than to an individual is, to my mind, of no significance. There is no principled basis on which a distinction can be drawn between rights which are peculiar to an individual plaintiff and those which inhere in a plaintiff as a member of a particular class. And that is so even if the members of that class cannot be identified with complete accuracy.

Although it would not be strictly accurate in this case to describe the respondent as being in a position of control, its relationship with the appellants is closely analogous to that which obtains where one person is in a position to control the exercise or enjoyment of a legal right by another person. In this case, the respondent knew that there was a class of persons who availed themselves of the right to sell potatoes in the Western Australian market and/or who used their property and equipment to produce potatoes for that market. It knew that those who did so would lose those rights or would have them impaired for a period of five years if bacterial wilt were discovered within 20 kilometres of the place or places in which their potatoes were grown, cleaned, washed, graded or packed. And the respondent knew or ought to have known that, if bacterial wilt were to be transmitted to potatoes grown within that 20 kilometre zone, those persons who grew potatoes for the Western Australian market or who used their land and equipment for that purpose were powerless to protect their own interests.

In my view, where a person knows or ought to know that his or her acts or omissions may cause the loss or impairment of legal rights possessed, enjoyed or exercised by another, whether as an individual or as a member of a class, and that that latter person is in no position to protect his or her own interests, there is a relationship such that the law should impose a duty of care on the former to take reasonable steps to avoid a foreseeable risk of economic loss resulting from the loss or impairment of those rights. Notice of contention

I agree with Gummow J, for the reasons that his Honour gives, that there is no substance in either of the points pressed by the respondent pursuant to its Notice of Contention. Loss or damage

The case was argued in this Court and in the Full Court solely on the question whether there was a special feature or relationship such that the respondent might be held liable to the appellants for pure economic loss. In the case of each appellant, however, liability is dependent on his, her or its having suffered damage, a matter on which the pleadings could have been more precise. This notwithstanding, Perre's Vineyards and each of the individual appellants had an interest in land and may have suffered loss in consequence of a diminution in its value following the detection of bacterial wilt on the Sparnon property. And on the assumption that Warruga Farms and the Rangara joint venture each owned potatoes which either could not be sold or could not be sold in Western Australia, Warruga Farms and, as a member of the joint venture, Rangara would seem to have lost some part of the value of those potatoes.

Perhaps, some or all of the appellants suffered damage other than that referred to above. Perhaps, also, the assumption that Warruga Farms and the Rangara joint venture each owned potatoes which could not be sold in Western Australia is not justified. It is, however, convenient to dispose of the appeal on the basis of that assumption, leaving it to the parties to amend the pleadings appropriately or to move to have the action struck out to the extent that an appellant has not clearly alleged loss or damage. Orders

The appeal should be allowed with costs. The orders of the Full Court with respect to the present appellants should be set aside and, in lieu thereof, the appeal to that Court should be allowed with costs and Order 5 of the Orders made by von Doussa J varied to refer to the second and third respondents, Orders 6 and 10 set aside and the matter remitted to a single judge for further hearing. The costs of the first hearing should abide the outcome of the further hearing. The parties should have liberty to apply to a single judge to vary Order 12 of the Orders of von Doussa J as may be appropriate.

McHUGH J. The principal question in this appeal is whether a supplier of diseased seed to South Australian potato growers owed a duty to prevent economic loss to other growers, whose farms were a few kilometres away, in circumstances where the second growers sold most of their crop in the Western Australian market, and the supplier knew that Western Australia was a lucrative market for South Australian growers, but prohibited the import of potatoes grown within 20 kilometres of a known outbreak of the disease during the last 5 years. The supplier also knew that the second growers' farms were within 20 kilometres of the growers to whom the seed had been supplied, although I do not regard that knowledge as a decisive factor in the case (it would have been enough that the supplier could have ascertained the identity of the second growers.)

The supplier was the respondent, Apand Pty Ltd ("Apand"); the growers to whom the seed was supplied were the Sparnons; and the second growers were the Perres, a convenient name for a group of companies and partnerships, linked by membership of the Perre family, who are the appellants.

The Sparnons and the Perres sued Apand for damages for negligence in the Federal Court of Australia which held that Apand owed a duty of care to Sparnons and had breached it but owed no duty to the Perres. The Full Court of the Federal Court upheld those findings .

In my opinion, the Federal Court erred in not finding that Apand had a duty to take reasonable care to protect the Perres from economic loss and in not finding that it breached that duty by supplying the seed, through a selling agent, to the Sparnons. The decision and the reasoning in Caltex Oil (Australia) Pty Ltd v The Dredge "Willemstad" have been criticised, most notably by the Judicial Committee of the Privy Council in Candlewood Navigation Corporation Ltd v Mitsui OSK Lines Ltd . But in my opinion the decision in Caltex was correct. Although the facts of the present case are very different from those present in Caltex , the reasons (with some modification) that led this Court in that case to hold that the defendant owed a duty to the plaintiff to protect it from economic loss, apply here. The losses suffered by the Perres were a reasonably foreseeable consequence of Apand's conduct in supplying the diseased seed; the Perres were members of a class whose members, whether numerous or not, were ascertainable by Apand; the Perres' business was vulnerably exposed to Apand's conduct because the Perres were not in a position to protect themselves against the effects of Apand's negligence apart from insurance (which is not a relevant factor); imposing the duty on Apand does not expose it to indeterminate liability although its liability may be large; imposing the duty does not unreasonably interfere with Apand's commercial freedom because it was already under a duty to the Sparnons to take reasonable care; and Apand knew of the risk to potato growers and the consequences of that risk occurring. The Factual and Legal Background

The Perres grew potatoes and had a potato processing plant on land a few kilometres from the Sparnon farm in South Australia. They exported much of their potato crop to Western Australia which was a lucrative market for South Australian potato growers. The Western Australians paid $670 a tonne - much more than could be achieved in the local market. During 1992, 79.2% of the Perres' crop was exported to Western Australia. The Perre interests

The "Perre interests" can be divided into four separate groups: Warruga Farms Pty Ltd

Warruga conducted three activities: (i) growing potatoes on land owned by some of the Perres; (ii) processing potatoes grown on that land and potatoes purchased from Rangara; (iii) exporting processed potatoes to Western Australia.

Warruga's shareholders were the natural persons who owned the land on which it operated (although Frank and Caterina Perre held their shares through Perre's Vineyards Pty Ltd). Warruga pleaded that it had suffered damage because it had lost the opportunity to pursue the Western Australian export market and had incurred costs in mitigating its damage. Rangara Joint Venture

Rangara was a joint venture between Rangara Pty Ltd (owned by Francesco and Maria Perre) and Pasquale and Grace Perre. The venturers grew potatoes on the Rangara land. They pleaded that they had suffered loss by losing those sales. Perre's Vineyards Pty Ltd

Perre's Vineyards Pty Ltd owned a processing facility and the land upon which it was situated. Its shareholders are Frank and Caterina Perre. Perre's Vineyards let the processing facility to Warruga. It pleaded that it had suffered loss by losing the benefit of its tenancy with Warruga and the opportunity to re-let the land for 5 years because it is used as a potato processing facility and is within 20 kilometres of a bacterial wilt outbreak. Perres - Natural Persons

The Perres were the owners of Warruga, Rangara and Perre's Vineyards. They pleaded that they had suffered loss because the Warruga land had diminished in value and they had lost money on the sale of the Rangara land.

For the purposes of trial, the parties assumed that the Perres had suffered loss as pleaded. The Regulation

Regulations made under the Plant Diseases Act 1914 (WA) restricted the importation of potatoes into Western Australia to those which complied with its conditions. Relevantly those conditions were that the potatoes had to be certified that they were: "(ii) grown on a property situated at least 20 kilometres from a known outbreak of the disease Bacterial Wilt detected within the last 5 years; (iii) not, unless otherwise approved by the Director General, harvested, cleaned, washed, graded or packed with equipment or in premises with or in which potatoes, grown within 20 kilometres of a known outbreak of the disease Bacterial Wilt detected within the last 5 years, have been handled; ... " The Saturna seed experiment

At all material times, Apand commanded approximately 60% of the Australian potato crisping industry and had numerous contracts for supply of potatoes with growers. It employed staff to liaise with the growers and to remain familiar with the industry in their areas.

In 1987, Apand imported Saturna potato seed from The Netherlands to assess its suitability as a winter crop and for crisping. After introducing the Saturna seed into the Victorian Department of Agriculture and Rural Affairs' seed certification scheme, it removed the seed from the scheme in order to have it "processed out of existence", ie grown for consumption rather than for seed. It decided to do this in the Koo Wee Rup swamp area in Victoria - an area which Apand knew was generally unsuitable for the production of seed owing to the high risk of disease, including bacterial wilt. In February 1991, after a meeting of company executives at Pakenham in Victoria, Apand decided to resume the trial of the Saturna seed and have it grown at various properties throughout Australia. For commercial reasons, it decided not to resume the certification process. Instead it used seed grown in Koo Wee Rup swamp area (the "Tymensen seed") to grow further seed for the trials. Trials were conducted with 26 growers in South Australia, Queensland, NSW and Victoria. Six of them were in South Australia.

To enable the trials to be conducted in South Australia, Apand supplied Tymensen seed to Virgara Brothers in Adelaide. Virgara Brothers supplied the seed to the Sparnons on 2 February 1992. They planted it in early March. The land on which the Perres grew and processed potatoes was within 20 kilometres of the Sparnons' property. In April, bacterial wilt was observed on that property, Virgara's property and Tymensen's property. As a result, on 28 May 1992 the Perres received a letter from the South Australian Department of Agriculture informing them that they could not supply potatoes to Western Australia for 5 years. None of the land owned and operated by the Perre interests was affected by bacterial wilt. The finding of negligence

The trial judge formulated the duty of care owed by Apand to the Sparnons as follows: "In my opinion the duty imposed by the relationship was to take all reasonable steps to ensure that seeds which Apand provided to its growers had not come from a source where there was a real risk which Apand knew about or should have foreseen that the seeds might be infected by pests and disease."

That formulation of the duty was approved by the Full Court which also approved the trial judge's finding of breach of duty, expressed in these terms: "[T]he Sparnons would succeed merely on proof, which I find is established, that Apand, with its intimate knowledge of the potato industry and the Koo Wee Rup swamp area, should have foreseen the risk, which was a significant one, that seed potatoes multiplied in the Koo Wee Rup swamp area could be infected with a pest or disease that could be transmitted to subsequent growers using the seed or its progeny."

The Full Court found it unnecessary: "to give detailed consideration to his Honour's more general finding of a breach by Apand of its duty of care by not foreseeing the risk that seed potatoes multiplied in the Koo Wee Rup swamp area could be infected by pests or disease that could be transmitted to subsequent growers using the seed or its progeny."

These findings of breach of duty are not easy to understand. A defendant does not breach its duty simply because it foresees a risk of harm. Foreseeability of harm is an element of breach. A defendant only breaches a duty of care when it both knows or ought to know of a risk of harm from doing or failing to do an act and does or fails to do that act and has reasonable means available to it of avoiding that harm. Here the breach of duty was the supplying of seed which Apand knew or ought to have known would cause harm to the property of the Sparnons and which it could have avoided by not supplying the seed. Apand's knowledge

Apand knew of the risks associated with bacterial wilt and the vulnerability of potato farmers to the effects of the disease. Its knowledge or means of knowledge can be inferred partly from correspondence with the Victorian Department of Agriculture and Rural Affairs and internal company memoranda and partly from its admissions.

The trial judge found that Apand knew: (i) that bacterial wilt was a potentially disastrous disease for potato growers; (ii) that WA was an export market for South Australian growers and that a regulation prohibited the import of potatoes grown within 20 kilometres of an outbreak of disease; (iii) that the Sparnons were located in a potato growing area and that Warruga was a grower within 20 kilometres of the Sparnons.

An internal memorandum from Apand's Supply Manager to the Manufacturing Managers of Queensland, NSW, South Australia and Western Australia dated 18 April 1990 showed that it knew that "Bacterial Wilt is a potentially serious and pernicious disease which can cause heavy losses to growers. Infected paddocks take up to 4-5 years to clear and is a costly process." Another internal memorandum from the National Potato Supply Manager to the Manufacturing Managers of Queensland, NSW, Western Australia and South Australia dated 28 June 1990 shows that Apand knew that "[t]he economic impact on a grower who has the disease on his farm can be disastrous - he is unable to continue to grow potatoes in that part of the farm and unable to sell his infected farm" and that "[t]he major cause of spread is through growers buying non certified seed." A letter written on 11 April 1990 by an officer of the Victorian Department of Agriculture and Rural Affairs to Apand's National Supply Manager had informed him that "[t]he main way of spreading 'Bacterial Wilt' is by planting infected seed potatoes which, sometimes depending on the seasonal conditions or time of harvest, may not show obvious visual disease symptoms." Duty of Care and Proximity The demise of proximity as a unifying theme

Where a defendant knows or ought reasonably to know that its conduct is likely to cause harm to the person or tangible property of the plaintiff unless it takes reasonable care to avoid that harm, the law will prima facie impose a duty on the defendant to take reasonable care to avoid the harm. Where the person or tangible property of the plaintiff is likely to be harmed by the conduct of the defendant, the common law has usually treated knowledge or reasonable foresight of harm as enough to impose a duty of care on the defendant. Where a person suffers pure economic loss, however, the law has not been so willing to impose a duty of care on the defendant. By pure economic loss, I mean loss which is not the result of injury to person or tangible property.

Until 1963, the so-called "exclusionary rule" prohibited the recovery of any damages for pure economic loss suffered as the result of the negligence of another. Yet even before then, courts recognised that the rigour of the rule, if it ever existed in a pristine form, often occasioned injustice to plaintiffs. To overcome those injustices, the courts allowed recovery in a number of exceptional situations . Denial of recovery for pure economic loss remains the rule, but, since Hedley Byrne & Co Ltd v Heller & Partners Ltd was decided in 1963, many exceptions to the rule have been recognised. However, they have been developed in a haphazard and ad hoc fashion with no single principle underlying them. Indeed, Professor Feldthusen denies that such a principle may be divined, although he concedes that all cases of pure economic loss share certain characteristics .

No doubt one important reason why courts have felt the necessity to distinguish between liability for harm resulting in pure economic loss and liability for harm to person or tangible property is that pure economic losses frequently result in mere transfers of wealth. The plaintiff's loss is the defendant's or a third party's gain. Harm to person or property, on the other hand, ordinarily involves a net loss to social wealth. Furthermore, the risk of sustaining economic loss is "a burden which is much more often and easily spread than the risk of physical damage." So even with the demise of the exclusionary rule, courts in most jurisdictions still require a plaintiff in a pure economic loss case to show some special reason why liability should be imposed on the defendant.

In Murphy v Brentwood District Council , Lord Oliver said: "The infliction of physical injury to the person or property of another universally requires to be justified. The causing of economic loss does not. If it is to be categorised as wrongful it is necessary to find some factor beyond the mere occurrence of the loss and the fact that its occurrence could be foreseen. Thus the categorisation of damage as economic serves at least the useful purpose of indicating that something more is required ... "

For some years in this Court, the "something more" was "proximity" which Deane J suggested in Jaensch v Coffey "involves the notion of nearness or closeness". However, this Court no longer sees proximity as the unifying criterion of duties of care . The reason that proximity can not be the touchstone of a duty of care is that it "is a category of indeterminate reference par excellence." The need for a new framework for determining the existence of a duty of care

Although proximity may no longer be the talisman for determining a duty of care, neither this Court nor the English courts - which have also rejected proximity as the duty of care determinant - have entirely abandoned the use of proximity as a factor in determining duty. Nor do courts generally now appear to accept the categories approach although individual judges have favoured it. Under that approach, courts asked whether the instant case fell into a category where the existence of a duty had been judicially recognised or, alternatively, could be regarded as an incremental development of one of those categories.

Members of this Court have said that the differences between proximity as a criterion and the incremental approach are to a great extent more apparent than real . But neither proximity nor the categories approach or any synthesis of them has gained the support of a majority of Justices of this Court. Indeed, since the fall of proximity, the Court has not made any authoritative statement as to what is to be the correct approach for determining the duty of care question . Perhaps none is possible. At all events, the differing views of the members of this Court in the present case suggest that the search for a unifying element may be a long one. The Caparo test

The continuing use of proximity as a duty indicator in England appears most clearly in Caparo Industries Plc v Dickman where the House of Lords proposed a three stage approach for determining duty. That approach has been adopted in this Court by Kirby J . Under the three stage approach, the plaintiff can establish a duty of care only when it can prove three matters: first, that the causing of the damage was reasonably foreseeable; second, that a relationship of "proximity" existed; third, that in all the circumstances of the case, it is fair, just and reasonable to impose a duty of care on the defendant. However, in Caparo Lord Bridge (with whose speech Lords Roskill and Ackner agreed) went on to say "that the concepts of proximity and fairness ... are not susceptible of any such precise definition as would be necessary to give them utility as practical tests, but amount in effect to little more than convenient labels to attach to the features of different specific situations which, on a detailed examination of all the circumstances, the law recognises pragmatically as giving rise to a duty of care of a given scope." The three stage approach formulated in Caparo was applied by the House of Lords in Marc Rich & Co v Bishop Rock Marine Co Ltd . In my respectful opinion, the Caparo formulation suffers from three defects.

First, proximity as the second stage in a three stage test has no more content than it did when it was used as the unifying criterion, a point which it is fair to say was recognised by Lord Bridge in Caparo . If the meaning of proximity is restricted to nearness or closeness, neither logic nor policy requires that it should always be elevated above other factors that are relevant in a particular case. I think that Dawson J was correct when he said in Hill v Van Erp that proximity is neither a necessary nor a sufficient criterion for the existence of a duty of care . Furthermore, proximity in the sense of nearness or closeness is hardly a useful concept in most cases of pure economic loss.

Second, there is a danger that the Caparo test will be used as the test of duty in every case where duty is in issue. That would deny the operation of the established categories and the certainty that they provide. Even at its zenith, proximity was a rationale to be applied in aid of the principled development of new categories . It was not meant to "invade" the existing categories and wreak havoc with accepted and unproblematic principles developed within those categories.

Third, almost everyone would agree that courts should not impose a duty of care on a person unless it is fair, just and reasonable to do so. But attractive as concepts of fairness and justice may be in appellate courts, in law reform commissions, in the academy and among legislators, in many cases they are of little use, if they are of any use at all, to the practitioners and trial judges who must apply the law to concrete facts arising from real life activities. While the training and background of judges may lead them to agree as to what is fair or just in many cases, there are just as many cases where using such concepts as the criteria for duty would mean that "each judge would have a distinct tribunal in his own breast, the decisions of which would be as irregular and uncertain and various as the minds and tempers of mankind". Lord Devlin was surely right when he said: "For a judge to decide fairly and convincingly every case that comes before him in the light only of his own sense of justice, he would have to be a superman. I doubt if there have ever been more than a handful of men on the Bench who could do it, though doubtless there are slightly more who think that they could."

Furthermore, when legislatures and courts formulate legal criteria by reference to indeterminate terms such as "fair", "just", "just and equitable" and "unconscionable", they inevitably extend the range of admissible evidentiary materials. Cases then take longer, are more expensive to try, and, because of the indeterminacy of such terms, settlement of cases is more difficult, practitioners often having widely differing views as to the result of cases if they are litigated. Bright line rules may be less than perfect because they are under-inclusive, but my impression is that most people who have been or are engaged in day-to-day practice of the law at the trial or advising stage prefer rules to indeterminate standards.

No doubt in some cases judges cannot escape applying notions of "current ideas of justice or morality" in determining the duty question, any more than they can escape them in determining whether the foreseeable consequences of negligence should be regarded as too remote for liability to be imposed . As Cooke P pointed out in South Pacific Manufacturing Co Ltd v NZ Security Consultants & Investigations Ltd "whatever formula be used, the outcome in a grey area case has to be determined by judicial judgment. Formulae can help to organise thinking but they cannot provide answers." But if negligence doctrine is to escape the charge of being riddled with indeterminacy, ideas of justice and morality should be invoked only as criteria of last resort when more concrete reasons, rules or principles fail to provide a persuasive answer to the problem. The Anns test

In Anns v Merton London Borough Council , the House of Lords formulated a two stage test for duty that is still applied in Canada and New Zealand although it had been rejected in this Court and now in England . Under the Anns test, the court will find that a duty existed if the defendant knew or ought to have known that its conduct might cause harm to the plaintiff and there is no policy reason for negating the existence of such a duty . To approach the duty question by holding that reasonable foreseeability of the relevant harm grounds a prima facie duty of care widens the scope of liability, probably enormously and of itself provides no principled formulation. To now adopt it in this country would also be contrary to the growing consensus, evidenced by recent cases, that, if there is to be an in-built bias in the determination of duty of care, particularly in cases of pure economic loss, it should be against imposing a duty of care. According to that consensus, any potential claimant in a novel situation should affirmatively demonstrate that its loss deserves vindication via the law of negligence . Precise legal rights

Nor do I think that this Court should accept that a defendant should owe a duty of care merely because its conduct may defeat or impair "a precise legal right" of the plaintiff in circumstances where the defendant is in a relationship with the plaintiff and in a position to control the enjoyment of that right. The strongest judicial support for imposing a duty in this situation is found in the judgment of Gaudron J in Hill v Van Erp .

To give meaning to "precise legal right" in this context, however, requires a definition of "right". It is one thing if it means only proprietary, equitable and contractual "rights". But the very "right" recognised in Hill v Van Erp suggests that "a precise legal right" includes a right in the sense of an extension of the private law closure rule - ie that anything which is not prohibited is permitted, and therefore a "right". Hill v Van Erp shows that it is wrong to say there can be no claim in negligence for pure economic loss unless there has been an infringement of an existing right or interest. That case decided that loss of a mere expectation interest is recoverable in damages in negligence. But that does not mean that a duty should arise whenever a precise legal right has been infringed or impaired even if the plaintiff must also prove that the defendant controlled their relationship. To impose duties of care in such situations would extend the liability of defendants, perhaps massively. Many such cases would be - as was Hill v Van Erp itself - simply cases of transfers of wealth. Unlike damage to person or tangible property, there would often be no net loss of social wealth. The plaintiff's loss would be the defendant's or a third party's gain. Future categories of liability may develop which are based on a more limited definition of "right". However, that is a matter that I need not pursue in this case. The Perres no doubt had a right to trade, and that is a right that in various circumstances the law will protect , but not by imposing duties of care on others simply because they are in a position to control the enjoyment of the plaintiff's right to trade.

So if proximity is not the unifying test for negligence, if the two stage and three stage tests are defective, if the "precise legal right" formula is unacceptable and if the categories and incremental approach is not accepted favourably by the majority of judges, is there any solution to the problems posed by the development of a tort of negligent economic loss? Or must we now accept that Hedley Byrne was a glorious mistake and retreat to the exclusionary rule of the common law?

The exclusionary rule is often justified on the ground that it is certain - which it certainly is. But its certainty is obtained by rejecting claims that most people would agree ought to sound in damages. As will appear, it is not necessary, in my view, to use the exclusionary rule to obtain stability and predictability in the law of negligently inflicted pure economic loss. Furthermore, in Caltex Oil (Australia) Pty Ltd v The Dredge "Willemstad" , this Court rejected the exclusionary rule, and nearly 25 years later there should be no turning back. It is true that in Candlewood Navigation Corporation Ltd v Mitsui OSK Lines Ltd the Privy Council said that the judgments in Caltex did not disclose "any single ratio decidendi" . That being so, their Lordships "concluded that they are entitled, and indeed bound, to reach their own decision without the assistance of any single ratio decidendi to be found in the Caltex case." Even accepting the correctness of their Lordships' comments, in Australia it is Caltex and not Candlewood which represents the law. Whatever else Caltex may have decided, it determined that Australia no longer adheres to the strict exclusionary rule with or without defined exceptions . The Need for Predictability

Law is one of the most important means by which a Western society remains socially cohesive while encouraging the autonomy of its individual members and the achieving of its social, political and economic goals. But the effectiveness of law as a social instrument is seriously diminished when legal practitioners believe they cannot confidently advise what the law is or how it applies to the diverse situations of everyday life or when the courts of justice are made effectively inaccessible by the cost of litigation. When legal practitioners are unable to predict the outcome of cases with a high degree of probability, the choice for litigants is to abandon or compromise their claims or defences or to expose themselves to the great expense and unpredictable risks of litigation.

The cost of litigation also increasingly denies effective access to, and competent representation in, the courts of justice. Most lawyers now charge hundreds of dollars an hour for their services and legal aid is often unavailable to litigants in tort cases. The cost of those services is substantially increased when lawyers cannot give advice to their clients without the need to read numerous and lengthy academic articles and judgments - this being one of them - to find out what the law is. The cost of those services is also substantially increased when trial lawyers have to make lengthy and complex arguments about what appellate courts have "decided" and what policies govern their cases. Inevitably, some litigants must compromise or abandon what they believe to be just claims or defences and be left with a sense of grievance. Many more litigants must question whether litigation is a rational course of action given the return for them after deducting irrecoverable solicitor and client costs and taking into account the risks always inherent in litigation.

Obscure rules or rules whose application cannot confidently be predicted also have a public cost. Cases take longer to hear both because time is taken up in debating what are the governing rules and how they apply and because obscure rules inevitably extend the range of materials which the judge will think are admissible because they may have relevance to the issues. Judges therefore hear fewer cases than they should with the result that more judges must be appointed or the administration of justice becomes beset with unreasonable delays.

If negligence law is to serve its principal purpose as an instrument of corrective justice, the principles and rules which govern claims in negligence must be as clear and as easy of application as is possible. Ideally, arguments about duty should take little time with need to refer to one or two cases only instead of the elaborate arguments now often heard, where many cases are cited and the argument takes days. The needs of the litigant or potential litigant, the legal practitioner and the trial judge should guide the formulation of the applicable principles. That does not mean, however, that the common law must adopt arbitrary "bright-line" rules for the sake of certainty at the expense of what most people including judges would regard as a desirable result.

In many areas of law, judges cannot realistically ignore the past or change legal rules. This is the case with constitutional law and statutes. In those areas, the doctrine of separation of powers requires that judges be faithful to the past decisions of legislators and the makers of the Constitution . But in the area of judge-made law, the duty of judges to be faithful to the past is weaker. While stare decisis is a sound policy because it promotes predictability of judicial decision and facilitates the giving of advice, it should not always trump the need for desirable change in the law . In developing the common law, judges must necessarily look to the present and to the future as well as to the past. The Further Development of the Common Law of Negligence

Having rejected arbitrary exclusions, proximity, impairment of precise legal rights and Anns and Caparo as suitable determinants of duty, where does one find a conceptual framework that will promote predictability and continuity and at the same time facilitate change in the law when it is needed? I think that the existing legal materials already contain part of the answer. We have the established categories, a considerable body of case law and the useful concept of reasonable foreseeability. If a case falls outside an established category, but the defendant should reasonably have foreseen that its conduct would cause harm to the plaintiff, we have only to ask whether the reasons that called for or denied a duty in other (usually similar) cases require the imposition of a duty in the instant case. No doubt that may sometimes mean that, whether or not a duty is imposed at a particular time, will depend on the extent to which the case law has progressed to that time. But that is the way of the common law, the judges preferring to go "from case to case, like the ancient Mediterranean mariners, hugging the coast from point to point, and avoiding the dangers of the open sea of system or science" . It is not an approach that appeals to grand theorists who prefer to decide cases by general principles applicable to all cases. But in an area of law such as awarding damages for negligently inflicted economic loss, which is still developing and which has been recently cast adrift from any unifying principle, there is no alternative to a cautious development of the law on a case by case basis . Perhaps another unifying principle may emerge and gain widespread acceptance. Past experience suggests that, if it does, its fall from favour will not be long in coming. Until a unifying principle again emerges, however, the best solution is to proceed incrementally from the established cases and principles. The incremental approach is the most satisfactory approach

In my view, given the needs of practitioners and trial judges, the most helpful approach to the duty problem is first to ascertain whether the case comes within an established category. If the answer is in the negative, the next question is, was the harm which the plaintiff suffered a reasonably foreseeable result of the defendant's acts or omissions? A negative answer will result in a finding of no duty. But a positive answer invites further inquiry and an examination of analogous cases where the courts have held that a duty does or does not exist . The law should be developed incrementally by reference to the reasons why the material facts in analogous cases did or did not found a duty and by reference to the few principles of general application that can be found in the duty cases.

Further, I think that, so far as possible, the reasons for upholding or denying a duty in particular cases should be regarded as principles to be applied in determining whether a duty exists in cases within that category. Such reasons will reflect policies that the courts have recognised as relevant in determining the duty issue. In some cases, they will be so decisive in determining duty that they can be applied as rules or principles in other cases.

The present case is not one falling within any categories of liability hitherto recognised. The Canadian Supreme Court has adopted the categorisation of cases of pure economic loss proposed by Professor Feldthusen : 1. The Independent Liability of Statutory Public Authorities; 2. Negligent Misrepresentation; 3. Negligent Performance of a Service; 4. Negligent Supply of Shoddy Goods or Structures; 5. Relational Economic Loss.

The present case is not within any of these categories. It is perhaps closest to the last category. Professor Feldthusen has said that relational economic loss exists "when the defendant damages property owned by a third party and the plaintiff thereby suffers economic loss because of some relationship that exists between the plaintiff and the third party." However, this case is not a typical case of relational economic loss as the relationship between the third party (the Sparnons) and the plaintiff (the Perres) is purely a matter of physical proximity.

In Bow Valley Husky (Bermuda) Ltd v Saint John Shipbuilding Ltd McLachlin J (La Forest J concurring) held that the categories where relational economic loss had been recovered in Canada were: "(1) cases where the claimant has a possessory or proprietary interest in the damaged property; (2) general average cases; and (3) cases where the relationship between the claimant and property owner constitutes a joint venture."

None of these three categories applies in this case. The present case is therefore novel in terms of the categories. But that does not mean that no duty of care was owed to the Perres. "The categories of negligence", said Lord Macmillan, "are never closed." The issue of duty must be decided by reference to the few general principles that appear to govern all cases of pure economic loss. The Reasons for Denying or Imposing a Duty of Care in Cases of Pure Economic Loss

In determining whether the defendant owed a duty of care to the plaintiff, the ultimate issue is always whether the defendant in pursuing a course of conduct that caused injury to the plaintiff, or failing to pursue a course of conduct which would have prevented injury to the plaintiff, should have had the interest or interests of the plaintiff in contemplation before he or she pursued or failed to pursue that course of conduct . That issue applies whether the damage suffered is injury to person or tangible property or pure economic loss . If the defendant should have had those interests in mind, the law will impose a duty of care. If not, the law will not impose a duty. Illustrations of situations where the law does not require a defendant to contemplate the interests of the plaintiff and refuses to impose a duty of care are situations where plaintiff and defendant are engaged in a joint illegal enterprise , where the defendant is a public body exercising a quasi-legislative function and where the defendant is the Crown conducting military operations against the enemy during wartime . For policy reasons, the law denies a duty in each of these situations, irrespective of whether the damage suffered is injury to person or tangible property or pure economic loss. For policy reasons, the law may also deny or restrict the imposition of a duty of care because of the kind of damage suffered. Thus, only in limited situations does a defendant owe a duty to take reasonable care to avoid nervous shock being suffered by another . Similarly, in many situations, a defendant owes no duty to avoid causing economic loss to another person.

Until 1963, the almost universal rule was that, absent contract or fiduciary relationship, a person owed no duty to avoid causing economic loss to another person, and, although no longer a universal rule, no duty is the general rule. Judges and academic commentators have subjected this exclusionary rule to an intense scrutiny which has yielded a broad consensus on the rationale for the rule. They generally agree that the theoretical underpinnings of the exclusionary rule are the need to avoid imposing indeterminate liability and the need to avoid imposing unreasonable burdens on the freedom of individuals to protect or pursue their own legitimate social and business interests without the need to be concerned with other persons' interests .

If the policy of the law is that indeterminacy of liability and conduct legitimately protecting or pursuing a person's social or business interests should not give rise to a duty of care, that policy should be translated into forms which can be applied as rules of law. I see no reason why that step should not be taken . However, indeterminacy and conduct legitimately protecting or pursuing a person's social or business interests are merely factors which negative the existence of a duty. That is an important limitation on their utility as a principle for determining whether a duty exists. Recognition of that limitation also answers the criticism that indeterminacy of liability and conduct legitimately protecting or pursuing a person's social or business interests are not useful criteria in determining duty because they are not relevant to all cases of pure economic loss. On the contrary, they are useful because, when they apply, they provide valid reasons for rejecting a duty. It hardly needs to be said that, when they are absent, no duty, or even a prima facie duty, automatically arises.

Nevertheless, when a court is satisfied that the economic loss suffered by the plaintiff was reasonably foreseeable by the defendant, that no question of indeterminacy of liability arises and that the defendant was not legitimately protecting or pursuing his or her social or business interests, it will often accord with community standards and the goals of negligence law, as an instrument of corrective justice, to hold that the defendant should have had the plaintiff's interests in mind when engaging or refusing to engage in a particular course of conduct. However, the common law in its desire to give effect to the autonomy of each individual does not generally require a person to act as if he or she were "my brother's keeper". That is particularly so when the defendant would have to take affirmative action to save a person from suffering harm.

What is likely to be decisive, and always of relevance, in determining whether a duty of care is owed is the answer to the question, "How vulnerable was the plaintiff to incurring loss by reason of the defendant's conduct?" So also is the actual knowledge of the defendant concerning that risk and its magnitude. If no question of indeterminate liability is present and the defendant, having no legitimate interest to pursue, is aware that his or her conduct will cause economic loss to persons who are not easily able to protect themselves against that loss, it seems to accord with current community standards in most, if not all, cases to require the defendant to have the interests of those persons in mind before he or she embarks on that conduct.

The principles concerned with reasonable foreseeability of loss, indeterminacy of liability, autonomy of the individual, vulnerability to risk and the defendant's knowledge of the risk and its magnitude are, I think, relevant in determining whether a duty exists in all cases of liability for pure economic loss. In particular cases, other policies and principles may guide and even determine the outcome. But I do not think that a duty can be held to exist in any case of pure economic loss without considering the effect of the application of these general principles. Indeterminacy

In Bryan v Maloney Mason CJ, Deane and Gaudron JJ pointed out that one reason why the law will often refuse to impose a duty to take reasonable care to protect another person from economic loss is its concern to avoid the imposition of liability "in an indeterminate amount for an indeterminate time to an indeterminate class." Concern about indeterminacy most frequently arises where the defendant could not determine how many claims might be brought against it or what the general nature of them might be. One feature that is more likely to be present in economic loss cases than physical damage cases is the "ripple effect" of careless conduct. Dr Jane Stapleton has pointed out that "economic loss can 'ripple' down a chain of parties; for example, the loss of profits which D causes P may in turn cause loss of profits to P's supplier and in turn to that supplier's suppliers, etc."

However, it is not the size or number of claims that is decisive in determining whether potential liability is so indeterminate that no duty of care is owed . Liability is indeterminate only when it cannot be realistically calculated. If both the likely number of claims and the nature of them can be reasonably calculated, it cannot be said that imposing a duty on the defendant will render that person liable "in an indeterminate amount for an indeterminate time to an indeterminate class."

That is not to say that the number of claims is always irrelevant in determining whether a duty of care is owed in cases of economic loss. In Caltex , Gibbs J thought that the size and number of claims were relevant. After giving an example of a case where the momentary inadvertence of the defendant had caused great loss to tens of thousands of people, his Honour said that "to require the wrongdoer to compensate all those who had suffered pecuniary loss would impose upon him a burden out of all proportion to his wrong." But even if this statement represents the current law - and it should be noted that courts do not hesitate to find a duty of care where an accident has caused extensive property damage or injury to many people - it is a policy of proportionality, not indeterminacy that prevents a court from imposing liability. The number of claims or their size, therefore, does not of itself raise any issue of indeterminacy. Indeterminacy depends upon what the defendant knew or ought to have known of the number of claimants and the nature of their likely claims, not the number or size of those claims.

In Caltex , where this Court held that a defendant, which had damaged a pipeline carrying oil, owed a duty of care to prevent economic loss to a plaintiff who relied on the pipeline to obtain oil, Mason J sought to overcome the problem of indeterminacy by holding that a duty would be owed only where the defendant "can reasonably foresee that a specific individual, as distinct from a general class of persons, will suffer financial loss as a consequence of his conduct." However, I see no reason to limit the existence of a duty to an "identified plaintiff". If the defendant knows or has the means to know who are the members of an ascertainable class affected by its conduct and the nature of the likely losses to members of that class, its liability is not indeterminate. In Caltex, Gibbs J obviously contemplated a wider test, saying : "The fact that the loss was foreseeable is not enough to make it recoverable. However, there are exceptional cases in which the defendant has knowledge or means of knowledge that the plaintiff individually, and not merely as a member of an unascertained class, will be likely to suffer economic loss as a consequence of his negligence, and owes the plaintiff a duty to take care not to cause him such damage by his negligent act." (emphasis added)

This passage indicates that his Honour would not confine the duty to persons whom the defendant knew would be affected by its conduct. His Honour would also extend the duty to at least a person who could be identified by using constructive knowledge. The judgment of Stephen J perhaps supports a wider approach that would include members of an ascertained class .

The common thread that may be drawn from the judgments of Gibbs, Stephen and Mason JJ in Caltex is that more than reasonable foreseeability of harm to a person is required before the defendant comes under a duty of care. Knowledge of harm to the plaintiff is a minimum requirement. However, in my opinion, the indeterminacy issue does not require that the defendant's knowledge be limited to individual persons who are known to be in danger of suffering harm from the defendant's conduct. Its liability can be determinate even when the duty is owed to those members of a specific class whose identity could have been ascertained by the defendant.

The problem of the "ripple effect" means that the courts must be careful in using constructive knowledge to extend the class to whom a duty is owed. It would not be wise, or perhaps even possible, to set out exhaustively when it would be permissible to rely on constructive knowledge. Speaking generally, however, it may be necessary to draw a distinction between using constructive knowledge to identify those within a class who are primarily affected by the defendant's negligence (the first line victims) and using constructive knowledge to identify those who have suffered economic loss purely as the result of economic loss to the first line victims. That is, as a general rule, no duty will be owed to those who suffer loss as part of a ripple effect. Ordinarily, it will be an artificial exercise to conclude that, before acting or failing to act, the defendant should have contemplated the interests of those persons who suffer loss because of the ripple effect of economic loss on the first line victims. While the defendant might reasonably foresee that the first line victims might have contractual and similar relationships with others, it would usually be stretching the concept of determinacy to hold that the defendant could have realistically calculated its liability to second line victims.

The reasoning in Caltex has been rejected by the Supreme Court of Canada and the Judicial Committee of the Privy Council . But I think that the decision was plainly right. The risk of loss to Caltex was reasonably foreseeable; no question of indeterminate liability arose; the defendant's freedom of action was not impaired by imposing a duty because it already owed a duty to the owner of the pipelines; Caltex could not readily avoid the risk of incurring the relevant expenses; and the defendant "must be taken to have known that carelessness in those operations, causing injury to the pipelines, would affect Caltex in precisely the way it did" . The only criticism that I have of the reasoning in Caltex is that it imposed too narrow a test for determining to whom a duty was owed. Unreasonable Burdens on the Autonomy of Individuals

One of the central tenets of the common law is that a person is legally responsible for his or her choices. It is a corollary of that responsibility that a person is entitled to make those choices for him or her self without unjustifiable interference from others. In other words, the common law regards individuals as autonomous beings entitled to make, but responsible for, their own choices. The legal doctrines of duress, undue influence and criminal liability are premised on that view of the common law. In any organised society, however, individuals cannot have complete autonomy, for the good government of a society is impossible unless the sovereign power in that society has power in various circumstances to coerce the citizen. Nevertheless, the common law has generally sought to interfere with the autonomy of individuals only to the extent necessary for the maintenance of society. In the law of liability for economic loss, we have a notable example of the common law's concern for the autonomy of individuals. In Hill v Van Erp , I pointed out that: "Anglo-Australian law has never accepted the proposition that a person owes a duty of care to another person merely because the first person knows that his or her careless act may cause economic loss to the latter person . Social and commercial life would be very different if it did. Indeed, leaving aside the intentional tort cases of wrongful interference with another person's legal rights (inducing breach of contract, intimidation and conspiracy, for example) a person will generally owe no duty to prevent economic loss to another person even though the first person intends to cause economic loss to another person. In our free enterprise society, no one questions the right of the trader to increase its advertising or cut its prices even though that action is done with the intention of taking the market share of its rivals."

The immunity from liability referred to in that passage is a consequence of the common law's concern for the autonomy of the individual and its desire to give effect to the choices of the individual by not burdening his or her freedom of action. Nor is the immunity confined to traders. As long as a person is legitimately protecting or pursuing his or her social or business interests, the common law will not require that person to be concerned with the effect of his or her conduct on the economic interests of other persons . And that is so even when that person knows that his or her actions will cause loss to a specific individual. Thus, a consumer owes no duty to a trader not to cause loss to that person by withdrawing custom. However, where other indicia of duty are present, the cloak of immunity cannot extend to conduct which cannot be fairly described as a legitimate pursuit or protection of a person's interests. What then is not a legitimate protection or pursuit of one's interests?

Competitive acts not prohibited by law are legitimate unless they fall within the ambit of one of the economic torts to which I referred in Hill v Van Erp . Ordinary competitive conduct imposes no duty to protect others from economic loss. At the other end of the spectrum, conduct involving deceit, duress or intentional acts prohibited by law could seldom, if ever, be regarded as done in the legitimate protection or pursuit of one's interests. However, it does not follow that, other indicia of duty being present, a person will always lose the immunity given to protect the autonomy of the individual merely because his or her conduct has been done in breach of law. It would be curious if breach of s 52 , or a provision of Pt IV , of the Trade Practices Act 1974 (Cth) automatically meant that the defendant owed a common law duty of care to all those that he or she knew would be affected by the breach. Between the extremes are acts whose legitimacy will no doubt affect minds differently. They are likely to involve sharp or ruthless conduct. Perhaps no more can be said in the abstract than that the line of legitimacy will be passed only when the conduct is such that the community cannot tolerate it.

Because protection of the individual's autonomy is the reason for the immunity conferred by the legitimate protection or pursuit of interest doctrine, that doctrine can have little, if any, application in a case where the defendant already owes a duty of care to do or not to do something. Thus, in Caltex , Stephen J, correctly in my opinion, thought a significant factor in support of the conclusion that Caltex was owed a duty of care was "the infliction of damage by the defendant to the property of a third party ... as a result of conduct in breach of a duty of care owed to that third party." Vulnerability

Cases where a plaintiff will fail to establish a duty of care in cases of pure economic loss are not limited to cases where imposing a duty of care would expose the defendant to indeterminate liability or interfere with its legitimate acts of trade. In many cases, there will be no sound reason for imposing a duty on the defendant to protect the plaintiff from economic loss where it was reasonably open to the plaintiff to take steps to protect itself. The vulnerability of the plaintiff to harm from the defendant's conduct is therefore ordinarily a prerequisite to imposing a duty. If the plaintiff has taken, or could have taken steps to protect itself from the defendant's conduct and was not induced by the defendant's conduct from taking such steps, there is no reason why the law should step in and impose a duty on the defendant to protect the plaintiff from the risk of pure economic loss.

In Esanda Finance Corporation Ltd v Peat Marwick Hungerfords , an important factor in denying a duty of care was that the plaintiffs were sophisticated investors well able in the circumstances to protect themselves. On the other hand, this Court found a duty in Hill v Van Erp and in Pyrenees Shire Council v Day partly because of the defendant's control (and knowledge) and relative inability of the plaintiffs to protect themselves. The law of contract

In determining whether the plaintiff was vulnerable, an important consideration will be whether the plaintiff could easily have protected itself against the risk of loss by protective action, particularly by obtaining contractual warranties. Pecuniary losses are one of the ordinary risks of business and, for that matter, ordinary life. Business people frequently take, or are easily able to take, steps to minimise their business or economic losses. Taking these steps will often be a more efficient way of dealing with the risk of these losses than requiring defendants to have regard to the risk that others may suffer economic loss. The economic efficiency of a society requires that the person best able to deal with or avoid the consequences of an economic risk from a cost view should be responsible for the risk and its consequences. In BDC Ltd v Hofstrand Farms Ltd , Estey J observed that "the realities of modern life must be reflected by the enunciation of a defined limit on liability capable of practical application, so that social and commercial life can go on unimpeded by a burden outweighing the benefit to the community of the neighbourhood historic principle." Where another body of law can effectively deal with economic loss, a court should be slow to use negligence law to impose a duty of care on a defendant. This is particularly important where to do so would interfere with a coherent body of law in another field.

In the twentieth century, many areas of economic activity are extensively regulated by legislation and regulations. The judgment of Brennan CJ in Pyrenees shows that the potential for interference with such a body of law is vitally important in determining whether a common law duty of care should be imposed on a defendant. In terms of pure economic loss generally, however, it is the relationship of negligence with contract that must ordinarily be considered. In Miller v United States Steel Corporation , Posner J, speaking of the term "economic loss", said : "It would be better to call it a 'commercial loss,' not only because personal injuries and especially property losses are economic losses, too - they destroy values which can be and are monetized - but also, and more important, because tort law is a superfluous and inapt tool for resolving purely commercial disputes. We have a body of law designed for such disputes. It is called contract law."

This passage contains an important truth in relation to cases of negligently inflicted pure economic loss. The agreement of the parties is ordinarily a better vehicle for determining their rights than imposing duties of care through the blunt application of negligence law. Nevertheless, while his Honour's assertion may (or may not) be accepted as a normative standard, it is not the law of Australia. To accept it at face value would mean that we should reject any exceptions to the exclusionary rule - including negligent misstatement and negligent performance of a service outside of contract. Furthermore, it overlooks that the economic torts and legislation, such as the Trade Practices Act 1974 , have already made substantial inroads into the natural domain of contract law. Australian courts must be careful before holding that the existence of obligations under a contract automatically denies liability in tort for pure economic loss . That said, if we are to aspire to a coherent law of civil obligations, courts must keep the contractual background in mind in determining whether a duty of care should be imposed on the defendant in pure economic loss cases. Developments in negligence should occur in sympathy with the law of contract. In Hill v Van Erp , Gummow J said: "Bingham LJ has observed that, like equity, the law of torts may, in appropriate circumstances, fill what otherwise are perceived to be 'gaps' in what should be one coherent system of law ... That, of course, is not to assert that the function of the law of tort, with respect to recovery of economic loss caused other than by reliance upon deceitful statements, is limited to the filling in of gaps left by the law of contract. But it is a starting point ..."

This Court has recognised that in certain circumstances concurrent duties in tort and contract can exist and that the law of contract and the law of negligence are informed by differing rationales . That difference supports the conclusion that the vulnerability of the plaintiff may often be a justifiable, but not sufficient, reason for imposing a duty of care in cases of negligence resulting in pure economic loss where the plaintiff could not have protected itself in contract. One of the assumptions of the law of contract, for example, is that the parties can bargain to protect their interests. A plaintiff who is vulnerable - for whatever reason - cannot do this in any meaningful way . In its quest for corrective justice, the law of negligence may be able to fill the gap which the law of contract has left.

Vulnerability will often include, but not be synonymous with, concepts of reliance and assumption of responsibility. The widely used concepts of "reasonable reliance" and "assumption of responsibility" have come under criticism . This Court has recognised that neither concept represents a necessary or a sufficient criterion for determination of a duty of care , saying that commonly, but not necessarily, a duty will arise in cases which "involve an identified element of known reliance (or dependence) or the assumption of responsibility or a combination of the two". This statement provides an insight into why both reliance and assumption of responsibility have been rejected as a unifying criterion in cases of pure economic loss. Like proximity, reliance and assumption of responsibility are neither necessary nor sufficient to found a duty of care.

In my view, reliance and assumption of responsibility are merely indicators of the plaintiff's vulnerability to harm from the defendant's conduct, and it is the concept of vulnerability rather than these evidentiary indicators which is the relevant criterion for determining whether a duty of care exists. The most explicit recognition of vulnerability as a possible common theme in cases of pure economic loss is found in the judgment of Toohey and Gaudron JJ in Esanda Finance Corporation Ltd v Peat Marwick Hungerfords .

Reliance may therefore be seen - for the purposes of duty of care - as an indicator of vulnerability: the plaintiff is specially vulnerable to the words and/or conduct of the defendant because he or she reasonably relied on the defendant. Reliance may also, of course, be relevant to causation. In terms of a duty of care, however, it is not reliance that is relevant, but its consequence, vulnerability. That is so, even though in certain situations "reasonable reliance" will be the appropriate test for determining whether the plaintiff was vulnerably exposed to harm from the defendant's acts or omissions.

The case law indicates that vulnerability as a test of duty is not restricted to the category of negligent misstatements. Nor are reliance and assumption of responsibility its only indicators. Thus, in Hill v Van Erp , a case of negligent performance of a service, which is closely analogous to negligent misstatement , neither reliance nor assumption of responsibility to the plaintiff was present . Justice Gaudron dismissed these criteria as indicators of duty and relied on the concept of control to found a duty . But that is simply another way of saying that the plaintiff is vulnerable to the defendant's conduct because the defendant controls the situation. In dissent in Hill v Van Erp , I said that the plaintiff's inability to protect her own interest (ie vulnerability) was the single strongest factor pointing to the existence of a duty . Brennan CJ, Dawson J and Toohey J based their judgments, at least in part, on a variant of the assumption of responsibility criterion which is a powerful indicator of the plaintiff's vulnerability to harm from the defendant's acts or omissions. Moreover, Dawson J and Toohey J also relied on the doctrine of "general reliance", which even more than "specific reliance" may be reduced to vulnerability .

In Pyrenees Shire Council v Day , Brennan CJ rejected general reliance as a useful concept in that case. Gummow J and Kirby J also rejected general reliance on the ground that it was a legal fiction and dependent on a principle - reliance - which already held too much sway in negligence . These criticisms of general reliance, however, do not apply to the concept of vulnerability. Vulnerability is not a legal fiction, nor is it dependent on reliance.

The degree and the nature of vulnerability sufficient to found a duty of care will no doubt vary from category to category and from case to case. Although each category will have to formulate a particular standard, the ultimate question will be one of fact. The defendant's control of the plaintiff's right, interest or expectation will be an important test for vulnerability. That test was applied by Gummow J in Pyrenees where his Honour noted that like the situation in Hill v Van Erp , there was no evidence of actual reliance . Insurance

Whether the plaintiff has purchased, or is able to purchase, insurance is, however, generally not relevant to the issue of vulnerability. In Esanda Finance Corporation Ltd v Peat Marwick Hungerfords , I pointed out that courts often wrongly assume that insurance is readily obtainable and that the increased cost of an extension of liability can be spread among customers by adding the cost of premiums to the costs of services or goods. In Caltex Stephen J rejected the contention that the existence of insurance or the more general concept of "loss spreading" were valid considerations in determining whether a duty of care existed . I agree with his Honour. They do not assist but rather impede the relevant inquiry . Loss spreading is not synonymous with economic efficiency - which will sometimes be a relevant factor in determining duty. Australian courts, however, have not accepted that loss spreading is the guiding rationale for the law of negligence or that it should be. Knowledge and Reasonable Foreseeability

The cases have recognised that knowledge, actual or constructive, of the defendant that its act will harm the plaintiff is virtually a prerequisite of a duty of care in cases of pure economic loss. Negligence at common law is still a fault based system . It would offend current community standards to impose liability on a defendant for acts or omissions which he or she could not apprehend would damage the interests of another. In Sinram v Pennsylvania Railroad Co , Learned Hand J said: "[S]o long as it is an element of imposed liability that the wrongdoer shall in some degree disregard the sufferer's interests, it can only be an anomaly, and indeed vindictive, to make him responsible to those whose interests he has not disregarded."

Conversely, where the defendant is not legitimately protecting or pursuing its own interests, current community standards would also seem to require that the knowledge of a defendant that its actions are likely to harm the interests of an ascertainable class of persons is a factor weighing in favour of imposing a duty. Questions of knowledge, actual and constructive, are always relevant in determining whether the defendant has breached a duty of care. But as the negligent misstatement cases show, they can be relevant in determining whether the defendant owed a duty to prevent economic loss to the plaintiff . To intrude questions of constructive knowledge of the risk into the duty question in every case of pure economic loss would run the risk of reinstating reasonable foreseeability as the criterion of duty in many cases . But where the defendant has actual knowledge of the risk and its consequences for an ascertainable class and is not legitimately pursuing or protecting its interests, I see no reason why that actual knowledge should not be an important factor in deciding the duty issue. Furthermore, because fault remains the basis of negligence liability, I see no reason why recklessness or gross carelessness should not be a relevant factor in determining whether a duty of care was owed. In Caltex , Stephen J seems to have thought that "the grossness of the wrongdoer's want of care" was a relevant matter in determining whether a duty of care should be imposed. Apand Owed a Duty of Care

Upon the facts of this case, whether or not Apand owed a duty of care depends on the answers to the following questions: 1. Was the loss suffered by the Perres or members of the group reasonably foreseeable? 2. If yes to question 1, would the imposition of a duty of care impose indeterminate liability on Apand? 3. If no to question 2, would the imposition of a duty of care impose an unreasonable burden on the autonomy of Apand? 4. If no to question 3, were the Perres or some of them vulnerable to loss from the conduct of Apand? 5. Did Apand know that its conduct could cause harm to individuals such as the Perres?

I do not think that any other factors are relevant in determining whether Apand owed a duty of care to the Perres. For example, no question of economic efficiency arises .

Apand does not dispute that the loss suffered by the Perres was reasonably foreseeable, and I think the rest of these questions should be answered favourably to the Perres or, at all events, to some of them.

Upon the facts of this case, therefore, Apand owed a duty of care to the Perres to protect them from the loss of sales to the Western Australian market and diminution in the value of their farming lands. That is to say, Apand owed the Perres a duty to protect them from pure economic loss. In this Court, counsel for the Perres argued that their potatoes had suffered physical damage in that they suffered partial economic immobility. This argument is totally devoid of merit. The physical attributes of the potatoes were not affected in any way whatever. The case is one of pure economic loss. The Full Court's Reasons

Fear that imposing a duty of care would involve Apand in indeterminate liability was the chief reason for the Full Court of the Federal Court rejecting the Perres' claim. The Full Court said : "When the decision in February 1991 was made to use the Saturna seed from the Tymensen property, it was subsequently supplied to some 21 growers throughout four States of Australia, including Virgara Bros in South Australia ... The Sparnon's property was but one of [the] 26 properties which planted second generation seed from the Tymensen property. No evidence was identified to this Court which indicated how many potato growers were within a 20 kilometre radius of each of those 26 ... growers, or even more widely, how many potato growers there were who had washed or packed their potatoes in premises in which other potatoes, grown within 20 kilometres of a known outbreak of bacterial wilt within the previous five years, had been handled. Nor was any evidence identified which would enable a finding to be made that some particular number or numbers of growers within that category did or might export potatoes to Western Australia ... ... There is no apparent reason why economic loss, due to inability to export potatoes to Western Australia, is of any special significance compared with other causes of economic loss; for example, a grower's extra expense in having potatoes washed or packed elsewhere ... ... At the material time, Apand had neither knowledge nor means of knowledge that the Perre interests individually, rather than as a member of an unascertained class, would be likely to suffer economic loss as a consequence of its negligence in supplying infected or potentially infected seed to the Sparnons. Even if one were to expand the permissible scope of knowledge or means of knowledge to a defined and small class of persons capable of ascertainment, we do not think the authorities presently indicate that that would be, of itself, sufficient, unless there were some other particular circumstances which drew to Apand's attention that it should, in taking the steps which it was found negligently to have taken qua the Sparnons, have also had regard to the interests of the members of that particular defined and small class of persons. There is no particular consideration of that nature in the present circumstances. Nor, in our view, does the evidence disclose anything in the nature of a common adventure ..."

It will be apparent from what I have written about indeterminacy that the Full Court has applied the wrong tests.

First, Apand need only have knowledge of an ascertainable class, not a "defined and small class". The size of the class is irrelevant. Consequently, evidence with respect to the number of growers that might possibly be affected was also irrelevant. All that is required is that at the time Apand supplied the seed to Virgara Brothers with the knowledge that it was to be supplied to the Sparnons, Apand knew that there was an ascertainable class whose members were at risk of economic loss. Its numbers are not to the point. The principle of indeterminacy is designed to protect the defendant against indeterminate liability, not numerous plaintiffs.

Second, the Full Court seemed to have accepted Apand's submissions that the relevant time for ascertainment of the class was at the Pakenham meeting. With respect, that was not the relevant time. No negligence arose until the time the infected seed was actually supplied to Virgara Brothers to be sold to the Sparnons, and the relevant time for determining whether it should have had the interests of the Perres in mind was when it decided to select the Sparnons as a grower. Apand's submissions on this point were designed to raise the spectre of indeterminate liability by presenting a scenario where growers all over Australia might be affected.

Third, the Full Court seems to have accepted Apand's submissions with respect to the second limb of the West Australian regulation regarding the washing and processing of potatoes - ie that the scope of sub-reg (iii) made the class affected effectively indeterminate, particularly because the class would not be closed at the time of breach. At the time of the supply of the seed to the Sparnons, Apand knew that bacterial wilt was a potentially disastrous disease for potato growers. It knew that it was a "notifiable" disease and that it was illegal to import infected seed into South Australia. It knew of the Western Australian regulation. It also knew that there was at least one potato grower within 20 kilometres of the Sparnons - Warruga. It follows therefore that Apand had actual knowledge of a class of persons defined by the scope of the regulation. It also knew that, if it was negligent, this class was likely to suffer economic loss in that the export business of members of the class to Western Australia would be devastated for 5 years. The class

The real issue on indeterminacy is whether the class was ascertainable or effectively indeterminate. It is clear from sub-reg (ii) that the ascertainable class includes growers exporting to Western Australia. The problem is sub-reg (iii) which states that no potatoes can be exported to Western Australia if they have been "harvested, cleaned, washed, graded or packed with equipment or in premises with or in which potatoes, grown within 20 kilometres of a known outbreak of the disease Bacterial Wilt detected within the last 5 years, have been handled". To allow liability for all parties falling within the scope of the second limb would be to allow liability to an indeterminate class. It would not have been easy for Apand to have ascertained the number of persons who might be affected by bacterial wilt on the Sparnons' land because they have businesses of harvesting, cleaning, washing, grading or packing within 20 kilometres of Sparnons. Moreover, the sub-regulation appears to include cleaners, washers, graders and packers of potatoes whose premises are outside the 20 kilometre area but who deal with potatoes grown within that area. It would be even more difficult for Apand to determine the nature of the claims that persons falling within sub-reg (iii) might make. That being so, a class of persons which included all those falling within sub-regs (ii) and (iii) should be regarded as giving rise to indeterminate liability. This is nevertheless not fatal to the Perre interests, or at least not to all of them.

For the purpose of this case, a more limited class is ascertainable and limited. That class should be defined as the owners of, and the growers of potatoes on, land within 20 kilometres of the Sparnons where potatoes grown on that land were exported to Western Australia. Whether or not Apand knew who were the members of the class is beside the point. They were easily identifiable, particularly by Apand. Indeed, it would be surprising if Apand did not know every member of the class. Moreover, the general nature of the likely claims of members of the class - loss of sales for at least 5 years and diminution in the value of land - was not so vague, particularly to a person in Apand's position, that Apand was at risk of an indeterminate liability in respect of an indeterminate amount.

However, I do not think that processors of potatoes (including any of the Perres) should be taken to be members of an ascertainable class. No doubt it was reasonably foreseeable that processors might be at risk because they were in a potato growing area and it was quite likely that some of them would process potatoes within the 20 kilometre sphere of risk. But some of them might easily be outside the area but process potatoes grown within the area. To include the processors within the class would be to engage in a very artificial process. It is logically impossible to include in the class those who processed potatoes produced within the 20 kilometre area or even those whose plant was within the area but to exclude from the class those who otherwise harvested, cleaned, washed, graded or packed potatoes with equipment or in premises within 20 kilometres of an infected property. That being so, the processors of potatoes were not members of an ascertainable class.

Applying the above definition of the class in this case, it is obvious that Warruga Farms is within it because it grew potatoes for export to Western Australia. However, I would exclude it in its capacity as a potato processor. That being so, its losses, in so far as they are attributable to its processing of potatoes, are not recoverable from Apand. Determining whether the other members of the Perre group are within the class is more difficult, but I am of the opinion that some of them are also within the class. Rangara Joint Venture sold its potatoes to Warruga. But it effectively grew potatoes for export to Western Australia. It should therefore be included in the class. Perre's Vineyards owned the processing facility but let it to Warruga. They are not within the class. As for the Perres themselves, in so far as they were growers or owners of land, they were within the class. Unreasonable burdens on the autonomy of Apand

What Apand did in this case was obviously in trade or commerce. It conducted the experiment to further its commercial interests - the production of potatoes that would be more suitable for winter growing and for crisping. It carried out the experiment in the way it did for its own commercial reasons. The question then is whether pursuit of those commercial interests makes the case one where Apand was under no duty of care to the Perres because it was legitimately pursuing its own business interests. In my opinion, that principle does not protect Apand in this case.

As I have pointed out, the immunity principle is the consequence of the common law's concern to protect the autonomy of the defendant and its freedom of action. Where the defendant is already under a restraint by reason of a duty owed to another, the rationale of the immunity rule disappears in respect of conduct within the ambit of that duty. Here Apand was already under a duty to take steps to protect the Sparnons against the consequences of bacterial wilt. That being so, Apand's autonomy and freedom of action is not relevantly impaired if it were held to owe the same duty to the Perres. The case would be one of extending liability for the same activity, not one of creating a new liability of a different kind with respect to different conduct.

Rejecting the claim for immunity based on the autonomy principle and the claim that to impose a duty would make its liability indeterminate does not mean, however, that Apand owed a duty to the Perres. It would have owed a duty only if there was some reason or reasons that required it to have had the interest or interests of the Perres in contemplation before supplying seed to Virgara Brothers for the purpose of Sparnons using it . In this case, those reasons must depend on the extent to which the Perres could easily protect themselves against economic loss and the actual knowledge of Apand as to the risks involved in the experiment and the magnitude of that risk. Vulnerability

At all times, Apand had control over the experiment and where it would occur. The relevant acts and omissions occurred outside the lands occupied by the Perres and without their knowledge. They were likely to suffer loss of great magnitude if the Sparnons' property was infected with bacterial wilt as a result of its negligence, as Apand well knew. It is indisputable that growers within the class, such as the Perres, were at great risk as the result of the supply of the seed. Ignorant of the danger developing at the Sparnons' property, the Perres were not in a position to do anything that could protect their economic interests. There were no contracts that they could enter into to protect themselves. No barriers or other protection could they erect to protect their land and crops. The future of their businesses depended on events some distance away of which they did not know and could not avoid even if they had known. The vulnerability of the Perres to a risk of such magnitude was very great.

But in addition Apand knew both the risks to the growers of potatoes and to the owners of land on which potatoes were grown and all the potential consequences for them in supplying uncertified seed to the Sparnons. It knew that "[t]he main way of spreading 'Bacterial Wilt' is by planting infected seed potatoes which, sometimes depending on the seasonal conditions or time of harvest, may not show obvious disease symptoms." It knew that "'Bacterial Wilt' is a potentially serious and pernicious disease which can cause heavy losses to growers" and that "[i]nfected paddocks take up to 4-5 years to clear". It knew that "[t]he economic impact on a grower who has the disease on his farm can be disastrous - he is unable to continue to grow potatoes in that part of the farm and unable to sell his infected farm" and that "[t]he major cause of spread is through growers buying non certified seed." It knew that the Saturna seed that the Sparnons were about to plant was uncertified.

In the light of its knowledge and the vulnerability of the Perres, the case for holding that Apand owed them a duty to take reasonable care to protect them from economic loss is overpowering. The law of negligence, as an instrument of corrective justice, would be in a sorry state if it did not require Apand to have regard to the interests of the Perres when it supplied the seed to Virgara Brothers for planting by the Sparnons. The Notice of Contention

I agree with the reasons of Gummow J for rejecting Apand's notice of contention. Order

I would make the following orders: 1. Appeal by the first to ninth appellants, and eleventh to fifteenth appellants, allowed with costs. 2. Appeal by the tenth appellant dismissed. 3. Set aside the orders of the Full Court of the Federal Court, and in lieu thereof order that the appeal to that Court: a) by the first to ninth appellants, and eleventh to fifteenth appellants, be allowed with costs; and b) by the tenth appellant, be dismissed. 4. Set aside paragraphs 5, 6 and 10 of the Order of von Doussa J made on 20 and 26 December 1996. 5. Remit the matter to a single judge of the Federal Court for further hearing conformably with these reasons, with costs of first hearing to abide the result of that hearing.

GUMMOW J. The appellants ("the Perres") bring this appeal against the rejection by the Full Court of the Federal Court (O'Loughlin, Branson and Mansfield JJ) of their appeal against the dismissal by a judge of that Court (von Doussa J) of their action against the respondent ("Apand") claiming damages for negligence. They did not pursue an appeal to the Full Court against the dismissal of their claims for damages for contravention by Apand of s 52 of the Trade Practices Act (Cth) ("the Trade Practices Act ").

The litigation arose from events which occurred in the course of 1991 and 1992. At that time, the Perres (the fourth to eighteenth applicants at first instance) and the Sparnons (the first to third applicants) carried on business as potato growers on properties situated between Berri and Loxton in the South Australian Riverland. There were over 150 potato growers in South Australia. It is unnecessary at this stage to indicate the particular activities of each of the Perres or the interrelation between those activities. In late April 1992, symptoms of an outbreak of bacterial wilt were detected in the Sparnons' experimental crop of Saturna potatoes. Apand had supplied to the Sparnons what it transpired were infected seeds under an arrangement whereby the Sparnons were to raise the experimental crop. At the time of the outbreak of bacterial wilt in 1992, Apand commanded 60 per cent of the potato crisping industry in Australia and it had contracted with potato growers in the eastern States and in South Australia for the supply to it of potatoes for processing as potato crisps. Apand's operations included research and development into new varieties of potato including potatoes which were small and round, with a high specific gravity and which could be grown as a winter crop. It was hoped that the Saturna potato would be shown to be one such variety.

Potatoes may be grown commercially either to produce commercial crops or to provide seed for planting out as a commercial crop. The Victorian Department of Agriculture and Rural Affairs had established a seed certification scheme ("the Certification Scheme"). Six generations of particular seed would be grown under departmental supervision by specialist growers. There would be a high expectation that certified potato seed which resulted would be disease free. Seed for the Saturna variety, which had been imported from the Netherlands in 1987, was grown for three generations under the Certification Scheme.

In mid-1990, Apand decided not to proceed with the development of Saturna because the tubers that had been grown had not been an appropriate size and shape. Third generation, non-certified, Saturna seed then was provided by Apand to be grown as a commercial crop. The cultivation took place on the property of Mr Tymensen which was in the Koo Wee Rup swamp area east of Melbourne. This has been an area of intensive potato growing for many years but is low lying and is susceptible to diseases. Mr Tymensen's property was not a certified seed growing area under the Certification Scheme. Mr Tymensen had been a grower for Apand for about 20 years but had never participated in the Certification Scheme. The particular land used for Saturna seed in the previous year had been planted with potatoes of a different variety and so was not "new" ground (ie, that not planted with potatoes for several years). Use of new ground would have minimised the risk of disease in the seed which was produced.

A meeting of Apand personnel was held at Pakenham in Victoria in February 1991 to decide the future of the Saturna variety and any use of the produce from Mr Tymensen's property. It was decided to proceed with field trials of Saturna as "a rapid seed multiplication programme". Later, in May or June 1991, Apand decided to conduct trials in South Australia. In December 1991, Apand invited five of its contract potato growers in different localities in South Australia, including the Sparnons, to grow experimental winter crops of Saturna potatoes. They each agreed and planted their experimental crops in February and March 1992 using Mr Tymensen's crop as seed. The purpose of the experiment was to investigate the suitability of the Saturna variety as a winter crop suitable for potato crisping by Apand. Apand agreed with the growers to purchase the crops so produced. Apand had already supplied Saturna seed to another South Australian grower, Virgara Bros, in July 1991. Virgara Bros now planted Saturna seed on their property at Virginia.

In April 1992, symptoms suggestive of bacterial wilt were discovered in all six crops and subsequent laboratory testing confirmed the presence of bacterial wilt in five of the six. Apand then discovered evidence of bacterial wilt in one corner of Mr Tymensen's paddock at Koo Wee Rup from which the seed had emanated. The primary judge found that, notwithstanding the confined nature of the infection, the harvesting and grading processes could have caused the disease to be distributed widely through other tubers and hence to the Sparnons' land.

There was no finding on the point, but it also should be noted that the importation of the seed into South Australia apparently had been forbidden by the Fruit and Plant Protection Act 1968 (SA) ("the Plant Protection Act") and delegated legislation thereunder . The presence of this regulatory regime, whether or not breach of it actually occurred, is a matter relevant to the existence and scope of a duty of care imposed upon Apand. However, whilst relevant, the presence of such a regime cannot be determinative of these common law questions. Legislative values may influence but do not necessarily dictate the content of the common law values which are in play. If this were not so, then there would be no recovery for negligently inflicted economic loss unless there was a breach of statute which justified placing the stigma of unacceptable business conduct upon the defendant's alleged tortious conduct. One might find it surprising that the events which gave rise to Caltex Oil (Australia) Pty Ltd v The Dredge "Willemstad" did not involve some infraction of statute. The pipelines, as Stephen J put it, "were apparently laid across Botany Bay ... pursuant to a licence granted by the public authority having control of the bay" . However that may be, liability in negligence was determined independently of any such considerations.

At the appeal in the Full Court, it was accepted that it was Mr Tymensen's property which was the source of the bacterial wilt which later appeared in the Sparnons' crop. Nor did counsel for Apand challenge in the Full Court the finding of von Doussa J that Apand owed a duty of care to the Sparnons arising from the relationship created when Apand invited the Sparnons to participate in growing an experimental crop with Saturna seed to be supplied by Apand.

There is no evidence that the potatoes grown by the Perres were blighted. However, they sold most of their produce, but not all of it, to buyers in Western Australia and did so at prices higher than could be obtained locally. The principal effect of legislation in Western Australia upon the Perres was to deny for five years access of their produce to that market by reason of its production within 20 kilometres of the outbreak on the Sparnons' property. The legislation adopted the 20 kilometre radius as a buffer or security zone against the risk of transmission of the disease.

The primary judge directed the determination at the same time of certain issues of liability in the actions brought by the Perres and the Sparnons separately from the remaining issues in their actions. The Sparnons made out their claims against Apand in negligence and in contract for breach of warranties of reasonable fitness and merchantable quality implied by s 14 of the Sale of Goods Act 1895 (SA) in the contract for the supply of the seed by Apand to the Sparnons. It is not clear whether the primary judge also based his conclusion, at least as to the warranty respecting reasonable fitness, upon the warranty implied by s 71 of the Trade Practices Act .

Von Doussa J entered judgment for the Sparnons against Apand for $51,200. However, as I have indicated, he dismissed the action against Apand by the Perres in respect of the causes of action for contravention of s 52 of the Trade Practices Act and for negligence, and the Perres appealed only in respect of the negligence claim. Apand appealed against the finding of liability in favour of the Sparnons. The Sparnons did not appear and the appeal proceeded in their absence. It was dismissed. "Pure economic loss"

On one branch of their argument, the Perres call in aid the reasoning of Jacobs J in Caltex Oil that the blockage of the processed crude oil through the pipeline, as a result of the negligent operation of the dredge, gave rise to a "physical effect [on the oil] short of physical injury" . By this means Jacobs J sought to confess but avoid the perceived bar against recovery of economic loss. However, the distinction applied by his Honour is illusory. It turns upon the notion of immobilisation of goods which are in the course of trade, an economic concept, and treats as physical damage the impediment to that trade.

The Perres' case is best approached on the substantial footing that they do not complain of "physical" damage to their land or the tangible assets used in their business operations there. They complain of what might be called "pure economic loss" sustained by reason of the outbreak on the nearby property having brought into immediate operation the Western Australian legislation which barred access to the lucrative market in that State. Apand supports the conclusion of the Full Court that, in those circumstances, there is no right of recovery.

The phrases "economic loss" and "pure economic loss" were used in argument and in many of the authorities to which the Court was referred. However, in Miller v United States Steel Corporation , Posner J, having spoken of the use of the term "economic loss" to distinguish it from an injury to the plaintiff's person or property, continued : "It would be better to call it a 'commercial loss,' not only because personal injuries and especially property losses are economic losses, too - they destroy values which can be and are monetized - but also, and more important, because tort law is a superfluous and inapt tool for resolving purely commercial disputes. We have a body of law designed for such disputes. It is called contract law." One may accept the first point made by his Honour without adopting the second. As is indicated in more detail later in these reasons, the so-called "economic torts" provide the legal means for the resolution of a range of commercial disputes which are beyond the reach of contract law. Nevertheless, in Miller , Posner J went on to refer, with evident approval, to the "increasing number of jurisdictions" in which it is held "that tort law provides no remedy in a case in which the plaintiff is seeking to recover for a commercial loss rather than damage to person, property, or reputation" . He added : "The insight behind the doctrine is that commercial disputes ought to be resolved according to the principles of commercial law rather than according to tort principles designed for accidents that cause personal injury or property damage."

In San Sebastian Pty Ltd v The Minister , Gibbs CJ, Mason, Wilson and Dawson JJ said : "The recovery of economic loss has traditionally excited an apprehension that it will give rise to indeterminate liability. And there is also an apprehension that the application of the standard of reasonable foreseeability may allow recovery of economic loss of such magnitude and in such circumstances as to provoke doubts about the justice of imposing liability for it on the defendant." Their Honours went on to point out that, although what had been said in Caltex Oil had been criticised, for example, in Candlewood Navigation Corporation Ltd v Mitsui OSK Lines Ltd , "the critics have themselves been unable to offer a solution to the problem" .

This attitude of the critics identified in San Sebastian and also of Posner J in Miller may reflect concerns that, unless there is denial of recovery for purely economic loss occasioned by the negligence of the defendant, there will be facilitated mass litigation, limitless liability and liability out of proportion to the fault of the particular defendant. A multiplicity of claims would be both vexatious to the courts (particularly in those jurisdictions where juries are still widely used in civil actions and there is limited scope for appellate or other review ) and unfair to the defendant whose careless slip may be completely out of proportion to the wide extent of the economic consequences. Enterprise may be discouraged and competition stifled. Further, many of the leading authorities in Britain and the United States involved admiralty and maritime law. In Canadian National Railway Co v Norsk Pacific Steamship Co , La Forest J pointed out that in large measure maritime law encompasses a global system and continued : "The bright line exclusionary rule against recovery has for nearly a century been in effect in that system, and continues to be followed by the major trading nations, in particular Great Britain and the United States. In making arrangements for allocating risks in essentially maritime matters, those engaged in navigating and shipping should, as much as possible, be governed by a uniform rule, so that they can plan their affairs ahead of time, whether by contract or insurance against possible contingencies."

However, the difficulty with the physical harm criterion remains that, while a lone act may cause finite physical harm, this nevertheless may be great and far-reaching in its consequences. Thus, it has been observed, with reference to the great fire of 1871, "[a] single overturned lantern may burn Chicago" . To give, or to continue to give, a general application of the bright exclusionary line identified by La Forest J may be to succumb, as Fullagar J put it, to "the temptation, which is so apt to assail us, to import a meretricious symmetry into the law" .

Moreover, the law of tort is concerned with more than the economically efficient allocation of risk and transfer of losses. In earlier times, strict liability in respect of invasion of the person or property of the plaintiff was the touchstone so that, as McHugh J put it, "the common law holds no prejudice against strict liability" . However, the general trend in modern authority, expressed in Northern Territory v Mengel , is "to the effect that liability in tort depends on either the intentional or the negligent infliction of harm" . In Caltex Oil , Stephen J identified, with reference to the speech of Lord Atkin in Donoghue v Stevenson , a broad principle underlying liability in negligence; the "general public sentiment" that, in the case at bar, there has been wrongdoing for which, in justice, the offender must pay. A similar point was made by McLachlin J in Canadian National Railway Co v Norsk Pacific Steamship Co . Her Ladyship said that the "contractual allocation of risk" argument overlooks : "the historical centrality of personal fault to our concept of negligence or 'delict' and the role this may have in curbing negligent conduct and thus limiting the harm done to innocent parties, not all of whom are large enterprises capable of maximizing their economic situation". McLachlin J also pointed out in the same passage that, in assigning liability to the "least-cost risk avoider", the argument assumes that (a) all persons or business entities organise their affairs in accordance with the laws of economic efficiency, and (b) all parties to a transaction share an equality of bargaining power which will result in effective allocation of risk. Moreover, it cannot be assumed that either all persons or business entities have sufficient information to assess risk or that the transaction costs in obtaining such risk information are equally burdensome.

The decision of this Court in Caltex Oil is authority at least for the proposition that, in a case such as the present, one does not begin with an absolute rule that damages in negligence are irrecoverable in respect of economic loss which is not consequential upon injury to person or property. The same may be said of more recent decisions of this Court and the House of Lords , as well as of the Supreme Court of Canada and the New Zealand Court of Appeal .

However, Apand founded upon the criticism of Caltex Oil in Candlewood Navigation Corporation , an appeal from the Supreme Court of New South Wales. The Privy Council there referred to the decision in Hedley Byrne & Co Ltd v Heller & Partners Ltd as having breached what otherwise should remain the limitation imposed by a rule that a plaintiff is not entitled to recover for economic or financial loss not consequential upon damage to his person or property. Their Lordships regarded the judgment of Blackburn J in Cattle v Stockton Waterworks Co as establishing such a rule "for reasons of practical policy" . Cattle also has been influential in various jurisdictions in the United States . The immediate result of Candlewood Navigation Corporation was that a claim for economic loss against a defendant who had damaged a ship operated by the plaintiff under a time charter failed because the time charter gave rise to contractual not proprietary rights.

Apand submitted that the present case did not fall within any category of recoverable economic loss. This submission assumes the correctness of taking the "rule" established by Cattle as a starting point, and is contrary to what is to be derived from Caltex Oil . Cattle v Stockton Waterworks Co

But does an examination of the state of affairs at the time of Cattle reveal that Blackburn J would have adopted a proposition of the width since attributed to him?

Certainly, even at the time Cattle was decided, there was no general prohibition upon the recovery of economic loss in tort. First, patents, copyrights and other forms of intellectual property which confer a permitted form of monopoly and are no more than incorporeal in nature, had long been protected by injunction and by actions for damages tried at law and treated as tortious . Secondly, even with respect to other property, such as interests in land, actionable damage might be purely economic in nature; the old action of slander of title in which the plaintiff complained that the defendant had scared off prospective purchasers of the plaintiff's property illustrates the point. Such decisions have been relied upon in the United States to support a principle of tort liability for interference with prospective advantage beyond that flowing from existing contractual relations . The present state of authority in a number of jurisdictions is discussed by the Supreme Court of California in Della Penna v Toyota Motor Sales, USA, Inc .

Thirdly, the economic loss inflicted upon the plaintiff by competitive but tortious conduct of the defendant may lie in the disruption or diversion of the commercial relations between the plaintiff and innumerable third parties. It is the apprehension of this occurring and the consequent damage to goodwill which attracts injunctive relief in passing-off actions. Moreover, since at least 1887 , it has been clear that it is sufficient for a plaintiff to show that the defendant dealt with the middlemen, such as wholesalers and retailers who are not deceived, and it is not necessary to show that the defendant dealt directly with consumers who are likely to be misled into believing that their goods in question emanated from the plaintiff.

Fourthly, Lumley v Gye established in 1853 that an action in tort lay against inducement by the defendant of breach of a contract for the provision of personal services to the plaintiff by a third party. In that case, no proprietary rights were involved and the loss was purely economic. Again, as the reasoning in Crofter Hand Woven Harris Tweed Co v Veitch shows, the object of a tortious conspiracy may be the infliction of injury to the economic interests of the plaintiff and it is those interests which the law protects. So, also, with the old action per quod servitium amisit , and the tort of misfeasance in public office which this Court recently considered in Sanders v Snell .

Thus, what often are identified as the "economic torts" provide many examples to support the proposition that Cattle cannot be taken as establishing any rule against recovery in tort of economic or financial loss unless it be consequential upon damage to the person or property of the plaintiff. In Attorney-General (Vict) v The Commonwealth , Windeyer J emphasised that : "The only reason for going back into the past is to come forward to the present, to help us to see more clearly the shape of the law of to-day by seeing how it took shape." Here, of course, the law is still taking shape. This makes the scrutiny of its antecedents all the more important, particularly when modern decisions such as Candlewood Navigation Corporation proceed upon assumptions as to what was settled a century ago.

It may be that most of the cases to which I have referred in examining the state of affairs at the time of Cattle can, at least in part, be explained as turning upon the necessity for malice or some other significant mental element, such as persistence after knowledge or notice of a plaintiff's rights. In Leigh and Sillavan Ltd v Aliakmon Shipping Co Ltd , Robert Goff LJ made the point that : "the philosophy of the market place presumes that it is lawful to gain profit by causing others economic loss, and that recognised wrongs involving interference with others' contracts are limited to specific intentional wrongs such as inducing a breach of contract or conspiracy. Certainly there seems to have developed an understanding that economic loss at the hands of others is something we have to accept without legal redress, unless caused by some specifically outlawed conduct such as fraud or duress; though how far this is the outcome of our reasoning, or the product of our law, is not altogether clear."

The determination of what, as an aspect of the economic organisation of society, is regarded as acceptable commercial dealing, is for the common law and statute. The debate in this area, in particular concerning the role of negligence, may turn upon an often unarticulated major premise that the common law values competitive conduct.

In Dorset Yacht Co Ltd v Home Office , Lord Reid and Lord Diplock referred to the attitude of the common law that competition involves one trader being entitled both to damage the interests of rivals by promoting that trader's interests and to withdraw custom although the supplier's goods or services have been entirely satisfactory. That is true as a general proposition but, as indicated above, requires qualification to accommodate the various intentional economic torts. The question is how far the law of negligence may develop without cutting too deeply into that common law value which favours competition. Writing extrajudicially, Holmes J said of that common law value : "Why is a man at liberty to set up a business which he knows will ruin his neighbor? It is because the public good is supposed to be best subserved by free competition. Obviously such judgments of relative importance may vary in different times and places." Any extension of liability in negligence should not be made without consideration of the point. For example, in the United States, Dean Prosser has favoured the view that, whilst liability for negligent rather than intentional interference with existing contractual relations or prospective economic advantage "is not impossible", such liability "must depend upon ... some special reason for finding a duty of care" .

In Australia, in developing the common law with respect to the limits of permissible competitive and business activity, this Court has had regard to what legislatures "have determined to be the appropriate balance between competing claims and policies" . In the United States, the existence of federal law upon such matters as intellectual property and the effect to be given to the Supremacy Clause may operate to restrict the development in the State courts of the common law of the State in question with respect to unfair competition . It remains to be seen whether any such "pre-emption" doctrine, operating beyond s 109 of the Constitution (which deals with conflict between federal and State laws), may apply in the development of the Australian common law respecting the "economic torts". Significantly, some of the anti-competitive practices forbidden by Pt IV of the Trade Practices Act (ss 45 -51AAA) do not require purposeful abuse of market power and instead fix upon a prescribed anti-competitive effect .

In Cattle it was held by the Court of Queen's Bench that the plaintiff contractor engaged to construct works on the land of a third person had no right of action after that land had been flooded by water leaking from a defective waterpipe laid by the defendant. This was so even though the result was that the plaintiff's contract with the landowner was rendered less profitable. As was pointed out in Caltex Oil by Gibbs J and by Mason J , Blackburn J may have based his judgment on the footing that the damage of which the plaintiff complained was not the proximate and direct consequence of the act of the defendant.

Speaking of the law as understood in 1952, Fullagar J, in Attorney-General for NSW v Perpetual Trustee Co (Ltd) , drew attention to the tendency to elide the distinction between duty and remoteness. His Honour said: "It cannot be doubted that the general rule of the common law is that which is stated in the judgment of Rich J in Quince's Case . 'The mere fact that the injury prevents a third party from getting a benefit from the person injured [which, but for the injury, he would have obtained ] does not invest the third party with a right of action against the wrongdoer' . It perhaps does not matter very much, for present purposes, whether we regard the rule as a rule relating to remoteness of damage, or whether we say that what it really means is that the wrongdoer has not been guilty of any breach of any duty owed by him to the third party. ... On the whole I would prefer to state the position in terms of duty, because it seems to me that, if the third party has a cause of action, it must rest on a distinct and different obligation from that subsisting between the wrongdoer and the person immediately injured."

It also should be remembered that it was not until comparatively recently that this Court determined that responsibility does not depend upon "the capacity of a reasonable man to foresee damage of a precise and particular character or upon his capacity to foresee the precise events leading to the damage complained of" and the Privy Council classified as sufficient a "real risk" which would not be brushed aside as far-fetched . The incautious application of criteria which stipulate "foreseeability" may, in economic loss cases, have the undesirable consequences to which Breyer J has pointed. In Barber Lines A/S v M/V Donau Maru , his Honour said : "To use the notion of 'foreseeability' that courts use in physical injury cases to separate the financially injured allowed to sue from the financially injured not allowed to sue would draw vast numbers of injured persons within the class of potential plaintiffs in even the most simple accident cases (unless it leads courts, unwarrantedly, to narrow the scope of 'foreseeability' as applied to persons suffering physical harm)." However, it needs to be kept in mind, particularly for the present appeal, that the criterion is " reasonable foreseeability". Liability is to be imposed for consequences which Apand, judged by the standard of the reasonable man, ought to have foreseen . Such a proposition was identified by Professor Stone as presenting a category of indeterminate reference . This, it should be accepted, allows in a given case, particularly one where the damage alleged was inflicted upon the economic interests of the plaintiff, for interaction between facts and values. Civil law

By way of comparison, it has been said of the position in France under the Code Civil that there: "(a) the limits on liability are provided by tests (certainty, indirectness, causation, assumption of risk) involving decisions of fact - albeit subject to the control of the Cour de Cassation - not black-and-white questions of law, and (b) not only is the outcome of a case not dependent by law on whether the loss is physical or economic, but even in similar fact situations the outcome may differ unpredictably from one case to the next, particularly when the case is being heard by a tribunal ". In Canadian National Railway Co v Norsk Pacific Steamship Co , Stevenson J noted that (contrary to what appeared to be the position in Germany) the civil law of Quebec and the civil law of France have no categorical rule preventing the recovery of pure economic loss. His Lordship added that, because of the experience of the Supreme Court of Canada in the civil law, "it cannot be frightened by apocalyptic scenarios about life after the rule against the recovery of pure economic loss" .

In his work, A Comparative Introduction to the German Law of Torts , Professor Markesinis refers to instances of recovery in Germany of economic loss which is immediately consequential to injury to person and property, and continues : "By contrast, when we move to pure economic loss ( reiner Vermögensschaden ) the difficulties experienced by German law are multiplied; in addition, they are experienced in areas which are well known to Common lawyers. Indeed, it could be argued that in no other area of its law of torts does German law demonstrate such an ideological affinity with the Common law as in its refusal to compensate pure economic loss through the medium of tort rules. Yet, as we shall note below, in both systems this basic premise has, in recent times, come under constant and ingenious attacks with practitioners (and judges) showing in many instances a willingness to probe for weak points in the citadel that would allow the establishment of bridge-heads for the opposing school of thought. Such an attitude, inevitably, gives rise to the question why should this be so; and the question becomes more pressing (and, to some extent has defied a conclusive answer) whenever the problem has been examined against the background of French law (and its derivatives) which have adopted the exact opposite solution, equated for all intents and purposes economic loss with physical injury, and have not incurred the dire consequences that Common lawyers and German lawyers so fear." The interests for which protection is sought

In Caltex Oil , Stephen J referred to the importance placed by the exclusionary rule as to economic loss upon the existence in the plaintiff of a proprietary or possessory interest in property. His Honour continued : "In the light of the origin, in the action on the case, of the tort of negligence, an origin in the context of which notions of the infringement of proprietary or possessory interests in property were by no means always essential to the cause of action, damage suffered being the gist of liability, it is curious that in this field of the tort of negligence an interest in property should be thought always to be a condition precedent to the right to recover for economic loss." It may be implicit in decisions such as Candlewood Navigation Corporation that whilst, as a general proposition, the integrity of person and property are sufficient interests for protection by tort law, they are necessary interests where "pure" economic loss is claimed and the defendant has not been fraudulent and otherwise lacks sufficient animus towards the plaintiff.

If that be the premise upon which the reasoning of the Privy Council in that case proceeds, then it should be rejected as inconsistent with Caltex Oil . The submissions by Apand stressed the attractions of Candlewood Navigation Corporation but did not seek to re-open Caltex Oil .

Further, there is much to be said for the point made by McPherson JA in Christopher v The Motor Vessel "Fiji Gas" as to the tendency of the common law, in attempting to define the limits of liability in negligence for economic loss, to concentrate first upon the actions of the defendant. Rather, there first should be identified those interests which are sufficient to attract the protection of the law in this field. Again, in Hawkins v Clayton , Gaudron J observed that : "where the act or omission complained of amounts to an interference with or impairment of an existing right which is known or ought to be known to the person whose acts or omissions are called into question then the issue of proximity may be open to determination by reference to factors somewhat different from those applicable where economic loss is occasioned without infringement or impairment of an otherwise recognized right." Her Honour added : "In actions in negligence for economic loss it will almost always be necessary to identify the interest said to have been infringed to determine whether the risk of loss or injury to that interest was reasonably foreseeable and whether a sufficient relationship of proximity referable to that interest was present so as to establish a duty of care."

One therefore begins the determination of the issues arising on this appeal by looking to identify the interests for which the Perres seek protection. These are not divergent but to a degree they are diverse.

Various members of the Perre family own what was identified as the Warruga land and the Rangara land. The properties shared a boundary. The Warruga land at its closest point was about 3 kilometres from the Sparnons' property and the Rangara land was about 2 kilometres distant. The Crown leasehold in the Warruga land was held by four Perre brothers and their wives, the first to eighth appellants. The Rangara land comprised two parcels of land, each owned by two Perre family members. A family company, Rangara Pty Ltd, the eleventh appellant, together with two family members, conducted a joint venture whose activities were the cultivation of potatoes and their sale to another family company, Warruga Farms Pty Ltd ("Warruga"). Warruga is the ninth appellant. That company, in addition to purchasing potatoes grown on the Rangara land, grew potatoes on the Warruga land and processed the potatoes from both sources. Warruga sold largely for export to Western Australia. It entered that market in 1990 and by 1992, of the produce sold, about 80 per cent went into the Western Australian market. A second family company, Perre's Vineyards Pty Ltd ("Vineyards") owned the processing facility and a leasehold interest in the land on which it was situated. Vineyards appears to have let the facility to Warruga as a tenant at will. Vineyards is the tenth appellant.

Warruga pleaded loss of export sales and expenses in mitigating its loss. Vineyards pleaded loss of its tenant and of the effective use of the facility for processing and packing. The owners of the Warruga land and the Rangara land pleaded respectively a decrease in value and loss on sale. The joint venturers alleged loss of income as a consequence of lost sales to Warruga.

The business activities of these parties depended in varying degrees upon the occupation and utilisation of land and the turning of the produce thereof to effective economic account. The substance of their complaints is that the continued pursuit of those activities was impeded significantly by the denial of access to the principal and lucrative market for that produce. There is debate as to whether each of the Perre interests can bring its complaint within such criteria and show the necessary foreseeability by Apand which would be "reasonable". However, the point of importance at this stage is that interests of the nature I have identified are susceptible of protection by the tort of negligence against injury, albeit economic in nature. Such injury is recognised as a kind of detriment which, if negligently caused, may attract compensation. No general formula

In Victoria Park Racing and Recreation Grounds Co Ltd v Taylor , Latham CJ suggested that the law might provide a remedy if there emerged some new method of interfering with the use and enjoyment of land and the suitability of the land for a particular commercial purpose. Given the context, his Honour had in mind a development in the tort of nuisance. No question of this arises immediately in the present case, particularly with respect to a liability to the Perres of the Sparnons for what was done on their land. However, it does suggest that the present may be a case of a "gap" in the common law and raise the issue of whether, consistently with principle, the tort of negligence may fill it .

Moreover, this case does not fall in a field where to allow recovery in negligence for economic loss would cut across a well developed body of doctrine which already applied, with its own checks and balances, to the situation in question. The wisdom of encouraging the further march of negligence across such fields remains a matter of debate. In Astley v Austrust Ltd , the Court recognised that "the law has evolved to the conclusion that concurrent liabilities in both contract and tort may arise in cases of professional negligence", but rejected the view that tort had superseded the implication by law of contractual terms. Moreover, in Downsview Nominees Ltd v First City Corporation Ltd , the Privy Council held that the principles of equity made provision as to the nature and extent of the duties owed by a mortgagee and a receiver and manager respectively to subsequent encumbrancees and the mortgagor, and that these principles were not to be replaced or supplemented by a liability in negligence .

The question in the present case is whether the salient features of the matter gave rise to a duty of care owed by Apand. In determining whether the relationship is so close that the duty of care arises, attention is to be paid to the particular connections between the parties. Hence what McHugh J has called the "inherent indeterminacy" of the law of negligence in relation to the recovery of damages for purely economic loss . There is no simple formula which can mask the necessity for examination of the particular facts. That this is so is not a problem to be solved; rather, as Priestley JA put it in Avenhouse v Hornsby Shire Council , "it is a situation to be recognised" .

The case law will advance from one precedent to the next. Yet the making of a new precedent will not be determined merely by seeking the comfort of an earlier decision of which the case at bar may be seen as an incremental development, with an analogy to an established category. Such a proposition, in terms used by McCarthy J in the Irish Supreme Court, "suffers from a temporal defect - that rights should be determined by the accident of birth" .

The emergence of a coherent body of precedents will be impeded, not assisted, by the imposition of a fixed system of categories in which damages in negligence for economic loss may be recovered. In Canada, it appears that what is identified as "contractual relational economic loss" is recoverable only in special circumstances and these can be defined by reference to three categories. The first is whether the claimant has a possessory or proprietary interest in the damaged property; the second is the general average cases in shipping law; and the third comprises cases where the relationship between the claimant and the property owner constitutes a joint venture. However, it has been necessary to add that the categories are not closed. The question in Bow Valley Husky (Bermuda) Ltd v Saint John Shipbuilding Ltd , where the case fell within none of the recognised categories, was whether the situation was one for the creation of a new category.

I prefer the approach taken by Stephen J in Caltex Oil . His Honour isolated a number of "salient features" which combined to constitute a sufficiently close relationship to give rise to a duty of care owed to Caltex for breach of which it might recover its purely economic loss . In Hill v Van Erp and Pyrenees Shire Council v Day , I favoured a similar approach, with allowance for the operation of appropriate "control mechanisms". In those two cases, the result was to sustain the existence of a duty of care.

In Esanda Finance Corporation Ltd v Peat Marwick Hungerfords , there had been no trial and thus no facts found. The pleading was bad because it did not allege facts adequate to carry the auditors into a sufficiently close relationship with the creditors or financiers of the company so as to found the element necessary to constitute a duty of care to the appellant. There, the potential for foreseeable but indeterminate and possibly ruinous loss by a large class of plaintiffs and other circumstances pertaining to the relationships between auditors, company and investors or creditors made it appropriate to take into account various "control mechanisms". For example, Toohey and Gaudron JJ pointed out that : "there is nothing to suggest Esanda was not itself able to have accountants undertake the same task on its behalf as a condition of its entertaining the possibility of entering into financial transactions with Excel. And, which is much the same thing in the circumstances of this case, there is nothing to suggest that it was reasonable for Esanda to act on the audited reports without further inquiry." The salient features of the present case

The Full Court referred to various matters in reaching its conclusion that the Perres had not shown the existence of the necessary duty of care owed to them by Apand. First, it was said that the Sparnons' property was but one of 26 growers throughout Australia (including nine in Queensland, six in Victoria and four in New South Wales) who had planted seed supplied by Apand from the Tymensen property, 18 of whom had produced crops infected by bacterial wilt. Then it was said that there was a lack of evidence to indicate such matters as the number of growers who were within a 20 kilometre radius of the properties of each of those 26 growers and who did or might export their potatoes to Western Australia.

This approach was adopted on the footing that the relevant decision which constituted the alleged negligent act of which the Perres complained was that taken in February 1991 at the Pakenham meeting, not the later and more specific decision to supply the Sparnons with the seed. At the time of the Pakenham meeting, no decision was taken to conduct trials in South Australia. As I have indicated earlier in these reasons, that decision followed in May or June 1991. Accordingly, it is difficult to see how any duty to any South Australian grower could have arisen or been broken at the time of the Pakenham meeting.

However, the primary judge held that Apand owed a duty of care to the Sparnons and that this arose when they were invited by Apand to participate in growing the experimental crop "with Saturna seed to be supplied by Apand". Further, von Doussa J held that the duty required Apand to take all reasonable steps to ensure that the seeds it supplied had not come from a source where there was a real risk, known to Apand or which it should have foreseen, that the seeds might be infected by pests or disease. The primary judge should have gone on to hold that that duty of care was owed to potato growers then conducting operations within a 20 kilometre radius of the Sparnons' property.

It was admitted on the pleadings that Apand knew or ought to have known that Warruga grew commercial potato crops on land located within the 20 kilometre buffer zone. The evidence suggested that there was only one other commercial potato grower conducting activities within the 20 kilometre zone. In any event, there was no suggestion that the growers conducting such activities in this area at the time of the supply of the seed to the Sparnons constituted a large category or a class the membership of which was not readily determinable by objective criteria. In any event, I would not accept that it necessarily is fatal in such a case that the members of the class to which the duty is owed cannot at all material times be identified with complete accuracy.

However, in this respect, Apand sought to derive comfort from the scope of the Western Australian law. This imposed a condition for the entry of potatoes into Western Australia. The condition required certification (in practice, supplied to the Perres by officers of the South Australian Department of Agriculture) that two negative criteria were met. The first was that, if the potatoes had not been grown under a certification scheme, they were grown, among other criteria, on a property situated at least 20 kilometres from "a known outbreak of the disease Bacterial Wilt detected within the last five years". The second was that, unless otherwise approved, the potatoes had not been: "harvested, cleaned, washed, graded or packed with equipment or in premises with or in which potatoes, grown within 20 km of a known outbreak of the disease Bacterial Wilt detected within the last five years, have been handled".

These criteria imposed an immediate burden upon a party seeking to enter potatoes upon the Western Australian market to provide certification as to the production history of the potatoes in question. Any duty of care owed by Apand in the circumstances of the present case certainly extended to Warruga which placed its potatoes into that stream of commerce.

It was suggested by Apand that the second criterion mentioned above could operate in respect of equipment which within the five year period passed from one owner or user to another and also would run with the ownership or control of the premises in which the potatoes were handled. It is unnecessary to determine whether, if the duty owed to the Perres extended to the processing facility owned by Vineyards, it would necessarily extend in favour of a class of users, owners or operators of equipment and premises which, whilst defined by the legislative criteria, was not a closed class at the time of the alleged breach of duty by Apand. It is sufficient to say that there would be difficulties in making out a case whereby liability was imposed upon Apand in favour of those further parties on the footing that Apand ought reasonably to have foreseen the consequences for them at future times and in future events.

The Full Court also appeared to impose upon the Perres an evidentiary burden to negative any prospect of Apand being rendered liable to an indeterminate class over an indeterminate period. There was no such burden upon the Perres.

The first point respecting the position of Apand is that, since at least 1990, it had appreciated the consequences of the spread of disease by the supply of uncertified seed in circumstances such as those in which it later supplied seed to the Sparnons. On 12 April 1990, Apand had received a copy of a letter written by the Victorian Department of Agriculture and Rural Affairs to a potato grower in the Gippsland area in which information and advice was given respecting "bacterial wilt". The letter included the following: "In the past month there have been two outbreaks of 'Bacterial Wilt' in potato crops at Buln Buln and Narracen. This is of grave concern as we have not had a serious outbreak of this disease for some years. In both instances the source of the disease can be traced to planting infected potatoes, that were sold as being 'good quality seed'. ... In both cases, the growers did not check adequately the background or history of the seed potatoes they purchased. Because the seed looked good it was accepted as being good quality. The main way of spreading 'Bacterial Wilt' is by planting infected seed potatoes which, sometimes depending on the seasonal conditions or time of harvest, may not show obvious visual disease symptoms. In addition to the immediate crop losses caused by the disease, there are two continuing affects [sic]: (1) Interstate and export sales of seed and ware potatoes. For example Western Australia prohibits the entry of potatoes grown within a 20 km radius of a Bacterial Wilt outbreak for a period of five years. ... The main points you should take to safeguard your farm and your markets are:- (a) Buy seed only from reputable people and ask about the history of the seed crop - be it certified or not. Certified seed is inspected by an unbiased agency, the Department of Agriculture and Rural Affairs (DARA), with experienced staff and must meet certain conditions before the crop is finally certified. There is not [sic] tolerance for 'Bacterial Wilt' in seed crops." (emphasis added) A week later, on 18 April 1990, Apand's Supply Manager wrote to other officers of the corporation, including the Manufacturing Managers in Queensland, New South Wales, South Australia and Western Australia, that bacterial wilt was a potentially serious and pernicious disease which could cause heavy loss to growers. He concluded: "The point of this note is to ensure that [Apand] does not become unwittingly a source for the spread of this serious disease. The company would be seriously embarrassed should such an event occur and become public knowledge." Here, Apand was proceeding on the assumption, no doubt correct, that its commercial interests militated against any involvement as an unwitting source for the spread of the disease. This is not a case where the common law would favour Apand's commercial interests by treating it as entitled to act so as to damage the commercial interests of others in the potato growing and processing industry. There is no question of inconsistency between the duty of care for which the Perres contend and what otherwise was the legitimate pursuit by Apand of its own interests . Indeed, as pointed out earlier in these reasons, it appears that the importation of the tainted seed from Victoria for the experiment using the Sparnons' land was forbidden under South Australian law, namely the Plant Protection Act.

Subsequently, on 26 June 1990, Apand's National Potato Supply Manager wrote to the Manufacturing Managers in the same four States to supplement the memorandum of 18 April. He said: "We can anticipate that should it be believed by growers that [Apand] was instrumental in the spread of the disease to their farm that litigation would be considered. [Apand] operations must implement adequate controls. Growers still complain of 'old' potatoes being returned in supposedly 'clean' bins. By way of example the Department of Agriculture and Field Officers in the USA where these diseases exist use disposable plastic overshoes etc. The Western Australia Dept of Ag will not allow potatoes sourced within 20 kms of a known outbreak of Bacterial Wilt into their state. The major cause of spread is through growers buying non certified seed. This often occurs where crops are grown out of season ie Winter harvest and regular certified seed is unavailable at that time of year. Hence there are major outbreaks throughout the Riverina of NSW, Maitland and now Southern Queensland. " (emphasis added) In addition, in 1990 Apand ceased its dealings with one of its growers in the Koo Wee Rup area, Mr Summers, after it had become aware that bacterial wilt had been detected on his property.

The result is that, at the time of the supply of the seed to the Sparnons, Apand knew or ought to have known that Warruga grew commercial potato crops within the 20 kilometre buffer zone, and knew of the special requirements of the Western Australian law with respect to importation of potatoes grown in adjoining areas to those where there was an outbreak of the disease. Apand was aware not only of the threat that bacterial wilt represented to growers, but of the presence of the disease in the Koo Wee Rup area where Mr Tymensen's property was located.

The Full Court found : "South Australia's export market in potatoes to Western Australia is very valuable. As a result, and to assist South Australian growers to meet the import requirements of the Western Australian authorities, the South Australian Department of Primary Industries involves itself in the examination and certification of potatoes intended for the Western Australian market." There was an issue before the primary judge as to whether at a time before the supply of the seed to the Sparnons Apand should be taken as having known that Warruga exported potatoes to Western Australia. It may be that a specific conversation between Mr Frank Perre and Mr Hughes, an employee of Apand, upon which the Perres relied in their case, involved no more than information casually given and is not to be treated as information, in the words of the trial judge, "conveyed to Apand for the purpose of its business". That point need not be decided. This is because the case is to be approached on the footing that, in an industry where the South Australian export market to an adjoining State was a valuable one and Apand was a very significant participant in the potato crisping industry in Australia, it at least ought to have known that there was a substantial possibility that growers within a 20 kilometre buffer zone of the Sparnons' property would be participants in that export market.

The evidence was that, leaving aside the Saturna seed, the only potato seed which Apand sent into South Australia had been certified under the Certification Scheme. Apand appreciated that the major cause of the spread of bacterial wilt was through growers buying non-certified seed. As a practical matter, Apand was in control of the initiation and conduct of the experimental activities using the Sparnons' property. Apand both selected and supplied the seed and chose the location for the experiment.

The Perres had no way of appreciating the existence of the risk to which they were exposed by the conduct of the Apand experiment and no avenue to protect themselves against that risk. They thus stood in quite a different position from that of the financier in Esanda Finance Corporation Ltd v Peat Marwick Hungerfords which had the power to deal from a position of strength in ordering its commercial relationship with the party to whom it provided financial accommodation . Here, the relevant risk to the commercial interests of the appellants was in the exclusive control of Apand. Its measure of control was at least as great as that of the Shire in Pyrenees Shire Council v Day . The effect of the Western Australian legislation upon the export market from South Australia was such that, within the buffer zone, the economic loss would flow directly and inevitably and the possibility of its occurrence would not be speculative.

The characteristics of the present case to which I have referred combined, subject to what follows, to bring the Perres and Apand into such close and direct relations as to give rise to a duty of care owed by Apand for breach of which purely economic loss may be recovered. The commercial enterprise and undertaking of the Perres was conducted through the medium of a complex of business structures, some of which dealt only or substantially inter se . Earlier in these reasons I have referred to the various components of the Perres' commercial enterprise and undertaking. Individual members of the Perre family, in different combinations, owned the Warruga land and the Rangara land, and these individuals are among the appellants. Warruga processed potatoes grown on the Warruga land and potatoes grown on the Rangara land which it had purchased from the joint venture and, in 1992, exported about 80 per cent of the processed potatoes to Western Australia. As I have indicated earlier in these reasons, the duty of care which I have identified was owed to Warruga.

The next question is whether the same is true of Vineyards. It owned the processing facility. The commercial utility of the facility was largely stultified by the operation of the Western Australian legislation. That result obtained because of the part the facility played in the overall Perre business. This was largely directed to the projection of its produce into the export trade, but not exclusively so. The facility remained available for use in domestic production, although that was less profitable to Warruga. The same is true of the Warruga land and the Rangara land. There is a question whether the duty extended to the individuals who owned these lands and to the members of the joint venture which grew potatoes for sale to Warruga. Apand was aware, as the letter received 12 April 1990 had made clear to it, of the threat to crops posed by the use of infected seed by "growers" but, in respect of the 20 kilometre buffer zone, the threat was perceived as one to interstate and export sales.

Vineyards and each of the individual appellants had interests in or related to land, the value of which was to some measure attached to the availability of the crops grown or processed by those appellants for projection into the course of the export trade to Western Australia. The joint venture would seem to have been at risk of a decline in the value of potatoes grown for supply to Warruga.

I would treat Apand's duty of care as extending to all these components of the Perre business and, on that footing, allow the appeal as to all of them. However, there remains the need on their part to show damage before they are able to demonstrate the existence of accrued causes of action. In that respect, the pleadings in their present state leave something to be desired. Further, in measuring the amount of damages recoverable, care will be required to ensure that the complex structure of the Perre business does not lead to double counting.

It is with these matters in mind that I would make the orders proposed by Gaudron J. The further hearing by a single judge will encompass the taking of such steps as the parties may be advised to put the pleadings in a satisfactory form, if this be possible, for the trial of the remaining issues in this action as to the sustaining of damage and the recovery of damages. Notice of Contention by Apand

At the hearing of the appeal, Apand advanced two grounds in its Notice of Contention. The first was that, if there be found to be any duty of care owed by it to any of the appellants, there was no breach thereof because its actions did not fall below the relevant standard of care. There is no substance in this ground. The standard of care exacted is that which is reasonable in the circumstances. The degree of care under that standard necessarily varies with the risk involved, including both the magnitude of the risk coming to pass and the seriousness of the potential damage that would follow .

Apand sought to diminish what otherwise one would have thought was the relevant standard of care in a case such as the present by reference to an alleged awareness on the part of the Sparnons that they were using non-certified seed. It is not easy to see what would be the relevance of that knowledge or awareness on the part of the Sparnons in measuring the standard required towards the Perres. Further, as indicated earlier in these reasons, the appeal by Apand against the Sparnons was dismissed by the Full Court. In any event, the primary judge found that: "[the Sparnons] knew Apand was heavily involved in research and development and promoted the use of good seed stock. To that end Apand encouraged and assisted its contract growers to use certified seed. I think it is highly improbable that the Sparnon partnership gave any thought to the possibility that experimental seed supplied to them by Apand could be diseased or could pose any threat to their property."

The second ground advanced by Apand is that a reasonable person in its position would not have foreseen that its conduct involved a risk of injury to the Perres or to a class of persons which included them.

The primary judge rejected the submission by Apand that it could not be liable because the occurrence of bacterial wilt involved a disease of a different type and kind from anything they could have foreseen as a consequence of the use of potatoes grown in the Koo Wee Rup swamp area as seed. However, reference has been made earlier in these reasons to the dealings between Apand and Mr Summers, a grower with property nearby that of Mr Tymensen, and the termination by Apand of those dealings after bacterial wilt was diagnosed on Mr Summers' property in 1990. His Honour found that in October 1990 there was a major break in the main drain that runs through the Koo Wee Rup swamp and a large volume of water was carried from Mr Summers' property to that of Mr Tymensen. His Honour concluded that, as a matter of probability the likely explanation for the bacterial wilt in Mr Tymensen's crop was that the flood in some manner carried it there from Mr Summers' property.

There was thus an ample basis for von Doussa J's finding: "that Apand, with its intimate knowledge of the potato industry and the Koo Wee Rup swamp area should have foreseen the risk, which was a significant one, that seed potatoes multiplied in the Koo Wee Rup swamp area could be infected with a pest or disease that could be transmitted to subsequent growers using the seed or its progeny" and that it was not: "realistic to attempt fine distinctions between types of pests and diseases, each of which can be transmitted by contaminated soil and potato tubers".

Neither of the two grounds on which the Notice of Contention was pressed should succeed. Orders

The appeal should be allowed with costs and consequential orders made as proposed by Gaudron J.

KIRBY J. This appeal takes the Court once again to the vexed question of the liability of an alleged tortfeasor in negligence for financial loss, unconnected with any injury done by the tortfeasor to the person or property of the plaintiff .

Lord Steyn has recently called this "the most controversial area of our law of tort" . Others, less charitable, have suggested that "[f]ew areas in modern tort law are darker and more uncertain" . Various epithets, ascending in anguish, have been applied to the efforts of the courts to express the applicable law. The decisions have been castigated as illogical, unjustifiable and historically suspect , "hopelessly deficient" , resulting from persistence with "formulaic" or "empty labels of no content" and designed to "mask" a judicial need for "control devices" which are not fully expressed. Only such devices, it is suggested, will keep in place the "floodgates" required to limit liability for economic loss and to provide legal practitioners with clear rules by which to advise their clients with confidence .

This last-mentioned consideration is one to which this Court must obviously attach importance . Claims in negligence for "pure" economic loss are increasing in Australia, as in other countries. Few such claims warrant appellate attention. It is therefore desirable that this Court should provide at least a framework or a methodology to guide practitioners and trial judges to a principled and predictable outcome in such claims. However, this is not a settled area of the common law , in Australia any more than in other jurisdictions. The law is in a state of development . This is unfortunate for everybody. But in default of legislation there is no escaping this fact.

It is not enough to say that this Court's approach, in any development of the ambit of liability, should be cautious and incremental . Of course it should. It is necessary to express the content of the approach which is proper and the criteria that will distinguish (so far as possible) a cautious increment that conforms to legal authority from an incautious one which would take the law beyond its acceptable boundary. That boundary is set, ultimately, by the answer to the question: ought the alleged tortfeasor to be under a legal obligation to observe care for the protection of the plaintiff against the incidence of the risk which has in fact ensued ? Inescapably, the answer to that question will reflect "a general public sentiment of moral wrongdoing for which the offender must pay" . But in the hope of affording better guidance, and a higher measure of predictability, than is offered by such a tautologous formula, the courts have developed first rules and categories and more recently approaches and methodologies to yield the law's response.

This appeal affords this Court an opportunity to clarify the law. Plainly it is an area of the law which calls out for such treatment. Only a measure of reconceptualisation will provide an enduring foundation for the application of legal principles to this and future cases in the place of the present disorder and confusion. The facts

The circumstances giving rise to the appeal, and the facts found by the primary judge , are set out in the reasons of the other members of this Court. I will not repeat the detail. Reduced to their essence they are relatively simple.

Apand Pty Ltd (the respondent), a major operator in the potato crisping industry in Australia, provided non-certified potato seed from a source in Victoria to a number of growers throughout Australia, including the Sparnons in South Australia. That seed produced a crop which was found to suffer from a disease known as bacterial wilt. The appellants comprise three distinct groupings of the Perre interests . They conducted their respective business operations in close physical proximity to, but not immediate contiguity with, the Sparnons. Relevantly, they were all within 20 kilometres of the Sparnons' property. The potato crops with which the Perre interests were concerned were not found to have become infected with bacterial wilt. However, a large part of the Perre interests' crop was ordinarily exported to the State of Western Australia. That State had a strict legal regime designed to quarantine it from any risk of the importation of bacterial wilt . Specifically, a certificate was necessary to permit the entry of potatoes into Western Australia. Such a certificate would not be granted in respect of potatoes which were grown within 20 kilometres of a known outbreak of bacterial wilt occurring in the previous five years, or which had been harvested, cleaned, washed, graded or packed using equipment or premises within the same radius. In consequence of the withdrawal of this certificate following the outbreak of bacterial wilt on the Sparnons' property, the Perre interests lost their substantial export market to Western Australia. They thereby suffered significant losses.

The Perre interests claimed damages from Apand, as did the Sparnons. The Perres and Sparnons also sued the State of South Australia for various suggested defaults in the State's policing of its own plant quarantine laws . The latter claims were dismissed by the primary judge. They were not pursued on appeal. But whereas the Sparnons succeeded at trial and on the appeal in their claim against Apand framed in negligence and contract , the Perre interests' claims in negligence were dismissed. A contention that their properties had in fact suffered physical damage within the authorities was not upheld. This left their claims as ones of "pure" economic loss. The Full Court of the Federal Court confirmed the primary judge's conclusion that the necessary relationship of proximity did not exist between Apand and any of the Perre interests with respect to the kind of damage suffered by them . It dismissed their appeals. It is from the rejection of their claims that the several Perre interests now appeal to this Court. The issues

Numerous factual disputes which occupied time in the Federal Court were resolved when the appeal by the Perre interests came for hearing. Apand no longer disputed that it owed the duty of care to the Sparnons found at trial and confirmed by the Full Court. Numerous factual obstacles which Apand had raised in the path of that finding must therefore be treated as resolved against Apand. These include the findings of the way in which, on the Tymensen property in Victoria from which the offending potato seed had been procured by Apand, bacterial wilt had come to infect the offending seed during a flood of which Apand was aware . They also include the finding of how the seed, allocated by Apand to South Australia, was allotted to Apand's contract potato growers in that State, including the Sparnons, and was in fact the source of the bacterial wilt which caused damage to them .

The damage to the Sparnons was the direct consequence of physical damage to their potato crop caused by the introduction of bacterial wilt onto their property. This was the same bacterial wilt that attracted the operation of the Western Australian law. It, in turn, effectively prevented the Perre interests for five years from exporting any of their crop to Western Australia. It had differential consequences for the several Perre interests to which it will be necessary to return.

There are various deficiencies in the way in which the claims for the Perre interests were pleaded. However, in so far as Apand's notice of contention seeks to re-canvass the findings made at trial, that its actions fell below the relevant standard of care or that damage to the Perre interests was unforeseeable to Apand, these contentions must be rejected. I agree in this regard with the reasons of Gummow J.

The quantification of damages suffered by the Perre interests was postponed whilst the issue of the liability of Apand to the Sparnons and to the Perre interests was separately tried. Accordingly, the remaining issue before this Court ultimately comes down to comparatively narrow, but still difficult, questions. These concern the liability of Apand under the common law of negligence to each of the Perre interests. More specifically, they concern whether, in the circumstances found (including those inherent in the now undisputed determination that Apand was liable to the Sparnons), Apand owed a duty of care to the Perre interests, or any of them. More specifically still, they concern the way in which the existence or absence of such a duty of care is to be decided in a case, such as the present, where the loss claimed is solely economic and where direct physical damage by Apand to the Perre interests' respective properties has not been shown.

If a duty of care to the Perre interests, or any of them, is established, the quantification of the consequences of the breach of that duty must be returned to the Federal Court for trial. The Perres submitted that the law did not produce the unjust and illogical result found below, namely that the Sparnons could recover for Apand's proved negligence damaging them but that the Perres, only a few kilometres away and being wholly innocent of the cause and damaged in a like way as a consequence of the same wrongful conduct, could not. Liability for pure economic loss

The common law originally forbade any recovery for financial loss in the absence of reasonably foreseeable harm causing damage to the plaintiff personally. This principle is often called "the exclusionary rule". Its origins go back at least to the second half of the nineteenth century in England. Although Lumley v Gye and Simpson v Thomson are sometimes mentioned, the usual starting point for tracing the rule is Cattle v Stockton Waterworks . There, the court dismissed a plaintiff's claim for financial loss suffered when the defendant negligently caused flooding of the land of a third party, thereby making more difficult and costly the construction of a tunnel on the third party's land which the plaintiff was contracted to build. Blackburn J, rejecting the claim, explained that the plaintiff could not recover the loss because the damage suffered was not a "proximate and direct [consequence] of [the defendant's] wrongful acts" .

The exclusionary principle was followed in a number of English decisions early in the present century . Gradually it evolved into a rule and was extended to a number of different fact situations . A similar rule was developed in the United States of America. In Robins Dry Dock & Repair Company v Flint , Holmes J observed: "[N]o authority need be cited to show that, as a general rule, at least, a tort to the person or property of one man does not make the tortfeasor liable to another merely because the injured person was under a contract with that other, unknown to the doer of the wrong. ... The law does not spread its protection so far." In a highly influential justification of the exclusionary rule, Cardozo CJ in Ultramares Corporation v Touche explained the common law's resistance to the imposition of such liability. It was undesirable, he said, to expose defendants to a potential liability "in an indeterminate amount for an indeterminate time to an indeterminate class".

The reformulation of the general law of negligence by Lord Atkin in Donoghue v Stevenson appeared to involve the potential to attach legal liability by reference to a duty concept expressed in the broad terms of foreseeability of harm. However, English decisions moved quickly to confine Lord Atkin's dictum to "negligence which results in danger to life, danger to limb, or danger to health" . The majority of the English Court of Appeal in Candler v Crane, Christmas & Co accepted that damage to tangible property could also attract Lord Atkin's neighbour test. But it said that it had "never been applied where the damage complained of was not physical" .

The illogicality and injustice occasioned by confining the law's protection by reference to the particular kind of damage suffered by the plaintiff was the foundation of the influential dissent of Denning LJ in Candler : "I can understand that in some cases of financial loss there may not be a sufficiently proximate relationship to give rise to a duty of care; but, if once the duty exists, I cannot think that liability depends on the nature of the damage."

In every jurisdiction of the common law, judges began to rebel against the injustice and apparently capricious illogicality of the exclusionary rule. "Exceptions" came to be recognised in the United States , England and Canada . As we can now see, the viability of the exclusionary rule was dealt a fatal blow (doubtless unintended) by the decision of the House of Lords in Hedley Byrne & Co Ltd v Heller & Partners Ltd . That decision overruled Candler. It upheld Denning LJ's minority conclusion. It decided that a duty of care to avoid pure financial loss could arise out of negligent advice upon which it was foreseeable that the plaintiff might rely. Lord Devlin administered the coup de grace to the old rule in these terms : "The interposition of the physical injury is said to make a difference of principle. I can find neither logic nor common sense in this. If irrespective of contract, a doctor negligently advises a patient that he can safely pursue his occupation and he cannot and the patient's health suffers and he loses his livelihood, the patient has a remedy. But if the doctor negligently advises him that he cannot safely pursue his occupation when in fact he can and he loses his livelihood, there is said to be no remedy. Unless, of course, the patient was a private patient and the doctor accepted half a guinea for his trouble: then the patient can recover all. I am bound to say, my Lords, that I think this to be nonsense. It is not the sort of nonsense that can arise even in the best system of law out of the need to draw nice distinctions between borderline cases. It arises, if it is the law, simply out of a refusal to make sense. The line is not drawn on any intelligible principle."

Since Hedley Byrne, in all jurisdictions of the common law, judges have conducted a search for an "intelligible principle" which could replace the discredited exclusionary rule. Putting it very generally, the options which have emerged are as follows: 1. To adhere to the exclusionary rule but to allow more "exceptions"; 2. To substitute an entirely new approach which has regard to all of the facts and to the considerations which support or resist the assignment of a legal duty of care, putting these considerations forward as if they were new rules or principles for determining the existence of a duty of care in the particular case; or 3. To reposition the ascertainment of the entitlement to recovery for pure economic loss within an established approach or methodology of general application such as that expressed in Anns v Merton London Borough Council or Caparo Industries Plc v Dickman . According to this option, it would be left to the criteria of "foreseeability", "proximity" and competing "policy" to provide the answer, in a particular case, as to whether or not a duty of care exists in law giving rise to an entitlement to recover economic loss. Indeterminacy of liability and the avoidance of interference with ordinary business conduct would be two of many policy considerations which would have to be weighed in adopting this approach. I shall consider each of these options in turn. Adhering to the exclusionary rule with exceptions

Speaking generally, until lately, the courts in England until lately and the courts of the United States have favoured the first option. Thus, in England, in Murphy v Brentwood District Council , the House of Lords confined the instances where the exclusionary rule would not apply to very narrow categories. Lord Keith of Kinkel pointed out that, in England, apart from the decision in Hedley Byrne and the narrowly decided holding in Morrison Steamship Co Ltd v Greystoke Castle (Cargo Owners) , no basis had been recognised to that time for the recovery of pure economic loss divorced from actual damage to the person or property of the plaintiff. Clearly, it was the fear of indeterminacy of liability which fortified their Lordships' conclusion in Murphy that a local authority was not liable for economic loss resulting from the negligent inspection of the foundation plans of a building development. In Henderson v Merrett Syndicates Ltd a similar approach was adopted. The Hedley Byrne principle was treated as an "exception" to the exclusionary rule which prevented a legal duty of care arising and thus recovery. Their Lordships in Murphy endorsed an approach earlier favoured by Brennan J in this Court, that it was preferable for the law to "develop novel categories of negligence incrementally and by analogy with established categories" rather than to adopt what their Lordships described as "a massive extension of a prima facie duty of care restrained only by indefinable 'considerations which ought to negative, or to reduce or limit the scope of the duty or the class of person to whom it is owed'" .

In Canada, the Supreme Court began to allow inroads into the exclusionary rule, as early as 1973 . However, a significant division of opinion about the proper approach to the problem emerged in Canadian National Railway Co v Norsk Pacific Steamship Co . That was a case in which a barge, being towed by a tugboat owned by Norsk, damaged a bridge with the consequence that the bridge was closed for several weeks. The closure resulted in economic loss to Canadian National Railway although none of its property was directly damaged. The Supreme Court of Canada was closely divided. The differences between La Forest J and McLachlin J illustrate the competition between those who would adhere to the exclusionary rule and allow categories of cases to form as exceptions, and those who would search for a new approach or methodology sufficient to put acceptable limits on liability, to discourage frivolous litigation, to promote commercial certainty but to permit the recovery of pure economic loss in cases considered appropriate .

La Forest J explained "the key difference" between himself and McLachlin J in Norsk as being "between a principled flexibility, which adheres to a general rule in the absence of policy reasons for excluding its application", and what he described as arbitrariness . Rejecting the proposition that "all economic loss cases are susceptible to the same analysis" , his Lordship embraced a five-fold categorisation of "exceptions" proposed by Professor Feldthusen . However, ultimately, La Forest J acknowledged that the reasons for supporting the continuance of the exclusionary rule were "essentially pragmatic" : "[D]enial of recovery places incentives on all parties to act in ways that will minimize overall losses, a legitimate and desirable goal for tort law in this area. Second, denial of recovery allows for only one party carrying insurance rather than both parties. Third, it will result in a great saving of judicial resources for cases in which more pressing concerns are put forward. The difficult job of drawing the line is at least done quickly without a great deal of factual investigation into the various factors that found proximity. ... Fourth, it also eliminates difficult problems of sharing an impecunious defendant's limited resources between relational claims and direct claims. Fifth, the traditional rule is certain, and although like any pragmatic solution, borderline cases may cause problems, the exceptions to the rule ... are reasonably well defined and circumscribed."

Similar reasoning is found in expositions in the United States supporting the exclusionary rule which still holds sway in that country. In State of Louisiana v M/V Testbank , Higginbotham J in the Court of Appeals (Fifth Circuit) explained why that Court was unpersuaded that it should "abandon physical damage to a proprietary interest as a prerequisite to recovery for economic loss" . Rejecting the suggestion that the judges should adopt a more "conceptually pure" substitute, his Honour said that the exclusionary rule possessed : "the virtue of predictability with the vice of creating results in cases at its edge that are said to be 'unjust' or 'unfair'. ... The answer is that when lines are drawn sufficiently sharp in their definitional edges to be reasonable and predictable, such differing results are the inevitable result - indeed, decisions are the desired product. But there is more. The line drawing sought by plaintiffs is no less arbitrary because the line drawing appears only in the outcome - as one claimant is found too remote and another is allowed to recover. The true difference is that plaintiffs' approach would mask the results. The present rule would be more candid, and in addition, by making results more predictable, serves a normative function. It operates as a rule of law and allows a court to adjudicate rather than manage."

The acknowledgment of "exceptions" scattered throughout the foregoing case law means that general adherence to the exclusionary rule could yet permit consideration of policy in the incremental development of new categories where recovery was permitted . Persistence with the exclusionary rule had some advantages. It promoted a measure of certainty and predictability of outcomes. It certainly helped to avoid indeterminate liability. It placed restraint on the law's interference with the economic freedom of actors in the marketplace. No doubt these and like considerations affected the decision of the Privy Council in one of the last Australian appeals which dealt with this issue.

In Candlewood , the Privy Council could find no clear ratio decidendi in the earlier decision of this Court in Caltex to justify a departure from the exclusionary rule. Their Lordships declined to follow Caltex . That case had held that the owners of a dredge were liable for the economic loss which was done to Caltex although no damage had been directly occasioned to Caltex's property. In Candlewood, the Privy Council applied the exclusionary rule. It denied recovery.

Persistence with the exclusionary rule, with the acknowledgment of exceptions to meet particular circumstances, is a viable approach for the law to adopt. It still has notable judicial supporters. However, that support has eroded in recent years. Courts have come to understand that it is essential to have a pre-existing conception of the basis for "exceptions" and especially as they multiply in number and kind. Stumbling along with more "exceptions", arising as the unidentified "special" features of the case demand the provision of recovery, would constitute "a striking acknowledgment of our inability to generate satisfactory general principles in this area" . It is not an approach which I regard as an acceptable foundation for the common law of Australia. New approach: allowing recovery where especially justified

The second option accepts that the exclusionary rule works unjust and illogical outcomes and cannot be sustained as a viable legal approach once important exceptions to it (such as Hedley Byrne ) are acknowledged. But this approach is equally unconvinced as to the utility of broad "tests" or criteria such as "foreseeability" or "proximity". Still less do considerations of justice, fairness and reasonableness (shorthand expressions for policy ) permit a line to be drawn, clearly distinguishing cases where legal liability exists from those where it does not.

Adherents to this second view regard the foregoing considerations as mere "labels" which add little to legal analysis . If this means that courts should recognise that the determination of liability depends on a pragmatic assessment, growing out of a thorough and detailed examination of all of the facts of the particular case, doing so without more would simply return the law to its basic question: ought there to be a legal obligation in the given facts ? Such an analysis would have the advantage of recognising more candidly that the underlying notion of liability in negligence at common law rests on a community sense of an obligation of one person to recompense another for wrongdoing . The best that could be done, upon this approach, would be to hold that courts, having abandoned the strict exclusionary rule, would draw upon "pockets of case law" for guidance from analogous cases decided in the past. Or they would compile a check-list from previous cases setting out the kinds of considerations which have tended to support the conclusion that a duty of care exists in law and those which point to the opposite conclusion .

I would not deny the influence of overall impression and assessment for differentiating the cases where recovery of pure economic loss is allowed from those where it is denied. Such assessments obviously lie behind the exceptions devised by the courts when the application of the exclusionary rule leads to results which are wholly unacceptable to the perceived justice of the common law.

However, there are two fundamental problems with the second option. In my opinion, they make it unviable. First, as a matter of legal authority, it does not grow out of past decisions in harmony with the methodology of the common law. Secondly, it provides no real guide for legal practitioners and courts whereby they can justify recovery in proper cases and avoid the potential of unlimited liability in others . Even if it is accepted that, to some extent, considerations such as "foreseeability" and "proximity" are "labels", masking deeper policy choices, they are rational labels. They are devised by the law to guide the decision-maker to the preconditions to the relationship between the parties which is necessary to the existence of a legal duty of care, and prerequisite to recovery. Even if the approaches adopted in Anns and Caparo , in their respective references to what is "just, fair and reasonable", give few precise tools to differentiate cases of recovery from those of non-recovery, they at least provide an approach or a methodology which the decision-maker can take. They may not constitute a "test" or "rule" as such. Lord Bridge in Caparo said they did not. I am prepared to agree. But they do provide an approach or methodology. That is certainly the way in which later decisions have reported them . It is an approach which obliges the decision-maker to face squarely the policy considerations which cannot be hidden behind a lawyer's conceit that liability in negligence for pure economic loss is an area of policy-free norms searching for a catalogue of "exceptions" to a very shaky general rule. As Cooke P remarked in South Pacific Manufacturing Co Ltd v New Zealand Security Consultants & Investigations Ltd : "There is no escape from the truth that, whatever formula be used, the outcome in a grey area case has to be determined by judicial judgment. Formulae can help to organise thinking but they cannot provide answers." The proper approach: foreseeability, proximity and policy

The foregoing analysis brings me to the third option. It was expressed in my reasons in Pyrenees Shire Council v Day . As an approach or methodology for deciding whether a legal duty of care in negligence exists, I suggested that the decision-maker must ask three questions: 1. Was it reasonably foreseeable to the alleged wrongdoer that particular conduct or an omission on its part would be likely to cause harm to persons who have suffered damage or a person in the same position? 2. Does there exist between the alleged wrongdoer and such person a relationship characterised by the law as one of "proximity" or "neighbourhood"? 3. If so, is it fair, just and reasonable that the law should impose a duty of a given scope upon the alleged wrongdoer for the benefit of such a person?

A study in the trends in the decisions of appellate courts in several countries of the common law suggests, to my mind, a gradual movement towards this approach, including in (but not limited to) cases of pure economic loss.

In Canada, the divided opinions in the Supreme Court in Norsk , to which I have referred, heralded an eventual transition to a new approach. In Norsk McLachlin J rejected the proposition that recovery could be determined by reference to pre-ordained categories as "exceptions" to the general exclusionary rule or by an appeal to an all-encompassing principle. Instead, her Ladyship approached the issue from the standpoint of a jurisdiction which still follows the two-step process for determining the existence of a duty of care in negligence as stated by the House of Lords in Anns . Under the rubric of policy McLachlin J referred to three specific considerations in that case : the need to place "some limits on the potentially unlimited liability which can theoretically flow from negligence" and to discourage "frivolous litigation"; the need for certainty "such that commercial enterprises have some appreciation of what risk is to be borne by whom" and the acceptance that, ultimately, recovery for economic loss is essentially a matter of policy. Try as judges might, they cannot devise hard and fast legal rules which will last long beyond the peculiar facts of the case in which they are formulated. This, with respect, is where I part company from McHugh J in this appeal. The rules which he has expressed are not universal. They are no more than criteria, applicable in the facts of this case, for giving content to the universal requirement of undertaking the policy analysis required by the third stage of the Caparo approach.

In Norsk , McLachlin J suggested that the Canadian decisions which had departed from the "narrow exclusionary rule" had not "opened the floodgates to claims for pure economic loss" . It was better, in her view, to recognise the necessity of policy choices than for courts to strain to allow or deny recovery "for the sake of doctrinal tidiness" . I agree.

Despite the differences in the Supreme Court of Canada in Norsk, the later decision of Hercules Managements Ltd v Ernst & Young saw, I suggest, a significant move towards the approach favoured by McLachlin J in Norsk. The opinion of the Court in Hercules was delivered by La Forest J, who had led the dissenters in Norsk. His Lordship expressed the opinion that the existence of a duty of care in cases of pure economic loss should be determined in the same way as in any other case of negligence. In Canada, that meant by the application of the Anns test . However, he acknowledged that the first limb of the Anns test (requiring that the plaintiff should demonstrate a relationship of proximity) was not very instructive. It was therefore necessary to "infuse that term with some meaning" . Relevant to the facts of the case in hand, his Lordship thought that the requisite degree of "proximity" would exist where a defendant ought reasonably to have foreseen that the plaintiff would rely on his or her representation and where reliance by the plaintiff would, in the particular circumstances of the case, be reasonable. In this way, although by a suggested elaboration of the notion of "proximity", the three elements which I consider to be critical were recognised in Canada: foreseeability, proximity and policy.

Most recently in Bow Valley Husky (Bermuda) Ltd v Saint John Shipbuilding Ltd the Supreme Court of Canada had before it another claim for recovery in negligence for pure economic loss. In that case a fire broke out on an oil rig causing it to be inoperative for several months. Two of the plaintiffs were oil explorers who had contracted for the rig to undertake exploratory work. McLachlin J returned to the approach that would achieve a just outcome whilst avoiding crippling liability which the application of the reasonable foreseeability rule alone had the potential to produce . Her Ladyship suggested that the differences between her approach and that of La Forest J in Norsk had been overstated. In practice, her tests of proximity and foreseeability, allied with considerations of policy, could ordinarily achieve the same result as the elaboration of categories of exception to the general exclusionary rule.

Once it was accepted that "new categories of recoverable contractual relational economic loss may be recognized where justified by policy considerations and required by justice" , it was necessary to appreciate that some concept must lie behind the recognition of a new category of exception and of the circumstances which would render it unjust to exclude recovery in the particular case. The same is true of a new set of rules or criteria called "principles" . Whilst Canada still nominally adheres to categories of exception to the exclusionary rule, it does so in a context of legal reasoning which is guided by the approach of Anns (with its frank reference to issues of policy) and a recognition that "the categories [of exception] are not closed" .

New Zealand courts have allowed recovery of pure economic loss in a variety of cases . They have approached the issue in terms of the principles ordinarily governing the ascertainment of a duty of care in negligence rather than as an aberrant branch of negligence law requiring complex, disparate and seemingly unconnected categories of "exception" to justify persistence with a general rule excluding recovery for pure economic loss.

In harmony with the Anns approach, the New Zealand courts have adopted a two-stage approach for ascertaining the existence of a duty of care in negligence. The first stage is concerned with the requirement of proximity ("which is not of course a simple question of foreseeability of harm as between the parties and involves the degree of analogy with cases in which duties are already established" ). The second stage asks whether there are policy considerations tending to negative or restrict the imposition of the duty. Cooke P expressed the opinion that policy had an inescapable part to play in determining whether a duty of care was imposed by the law in such cases. In his view, the notion of an "incremental" recognition of new classes or categories of liability by analogy with nothing more than existing forms represented a somewhat empty concept . Whilst accepting that the law's protection of personal bodily integrity over proprietary and economic interests was well grounded in legal history and policy, Cooke P concluded that the sharp distinction which the exclusionary rule required between pure economic loss and loss arising from damage to the person or property of the plaintiff was both arbitrary and spurious .

In Invercargill City Council v Hamlin Cooke P noticed that the general approach of the Canadian Supreme Court appeared to be moving in the same direction as that followed by the New Zealand courts. The courts in both countries recognised that "formulae and doctrine do not provide answers to new duty of care questions. In the end it is a matter of judicial judgment, formed after looking at established signposts and analogies." Although concern has been expressed by some writers that the New Zealand approach is insufficiently sensitive to the dangers of indeterminate liability , particularly in cases involving pure economic loss, the candid recognition of the role of policy in limiting or restricting the imposition of a legal duty of care affords the means, reasoning by analogy from past cases, to exclude unwarranted recovery in undeserving or inappropriate circumstances.

Even in England, since Hedley Byrne, an increasing number of judges have begun to reconceptualise negligence claims involving pure economic loss in terms of the general approach established by Caparo which I followed in Pyrenees . Thus in Spring v Guardian Assurance Plc Lord Lowry considered the question of whether a legal duty of care existed in negligence in that case by reference to the foreseeability and proximity tests. He then rejected the defendant's argument that public policy required that recovery be denied : "This argument falls to be considered on the assumption that, but for the overriding effect of public policy, a plaintiff who is in the necessary proximate relation to a defendant will be entitled to succeed in negligence if he proves his case. To assess the validity of the argument entails not the resolution of a point of law but a balancing of moral and practical arguments. This exercise could no doubt produce different answers but, for my own part, I come down decisively on the side of the plaintiff."

Spring involved an employment reference alleged to have been negligently prepared. Relevant to Lord Lowry's evaluation of the policy considerations applicable in that case, was the near certainty (as he found) that individuals would suffer damage as a result of such a reference and that a person "may be irretrievably blighted" as a consequence. These considerations, rather than the fanciful prospect that employers would be deterred from giving candid references or any references at all, carried the day for him. Lord Slynn of Hadley, having found that the torts of defamation and injurious falsehood did not, of their nature, necessarily exclude recognition of a duty of care in negligence, also approached the ascertainment of that duty in accordance with Caparo. In evaluating the third (policy) consideration, Lord Slynn was not swayed by predictions that recognition of a legal duty of care would unreasonably increase exposure to liability or inappropriately circumscribe freedom of speech or the provision of employment references .

Thus in these speeches in Spring, at least, the three-fold approach of Caparo provided their Lordships with a methodology for answering the question whether, in the facts found, a duty of care in negligence existed in law. The same approach has been adopted in other recent cases . This trend of authority should scarcely be surprising. If the three-fold approach is a proper one for the ascertainment of the general duty of care in negligence, despite suggestions that cases of pure economic loss justified different rules, it would (as Denning LJ remarked half a century earlier) be surprising indeed if a different approach and methodology (or entirely different categories, exceptions and rules) would be applied, simply because of the different "nature of the damage" .

The French law of civil obligations draws no distinction between physical and economic damage as such . Of course, French judges are as aware of the dangers of indeterminate liability as are judges of the common law . They too have sought to adopt control devices to limit the ambit of legal liability. One is the requirement that the injury suffered must have a "personal character" . In Germany, the law of civil obligations originally developed along lines not dissimilar to the common law's exclusionary rule . To some extent, that approach appears to have been reinforced by developments in German constitutional law . Nevertheless, cases do exist in Germany where courts have found a duty to compensate third parties for purely economic loss sustained as a result of an injury to another . In some cases, negligently or unreasonably inflicted commercial loss has been held actionable . So has economic loss intentionally caused by conduct which "shocks all fair-minded and right-thinking persons" .

A glance at civil law systems demonstrates a struggle, parallel to that in common law jurisdictions, to find a stable basis upon which to assign legal liability which will be determinate, predictable and not economically crushing and which will avoid unacceptable inhibition of the operations of the economic market. Because no simple solution may be derived from jurisdictions outside the common law which we could adopt as our own, it is necessary to return to the traditional approaches which the common law has taken to the tort of negligence generally. This does not mean continued persistence with the exclusionary rule, now thoroughly discredited and seen as illogical and unjust. But neither does it require persistence with the unsatisfactory progeny of that rule - the so-called "exceptions" which collect and multiply. Such exceptions wander on the face of the law in search of an unexpressed general principle. An equal error would be to pretend that there are brand new "principles". Fashioned to meet the needs of a given case, they would survive a year or two until seen for what they are - expressions of the considerations relevant to the policy choice to be made. Whether it is unpalatable or not, the policy choice cannot be avoided. Past decisions of this Court

In Candlewood , the Privy Council examined the reasons in this Court in Caltex which the trial judge (Yeldham J) had purported to apply. As I have said, their Lordships were "not ... able to extract from them any single ratio decidendi" . For that reason, and because they felt the separate expositions in Caltex incapable of offering "a satisfactory and reasonably certain guide" , they felt themselves free, and obliged, to return to the exclusionary rule expressed in the early English authority from the time of Cattle .

Now, it is true that the reasons of each of the participating Justices in Caltex were different from one another. However, the one common opinion was that an absolute application of the exclusionary rule was not acceptable as an expression of the Australian law of negligence . There were variations as to the extent to which the members of this Court were prepared to venture far from the harbour in which that rule still applied. Thus Gibbs J recognised the exclusionary rule as still applicable "as a general rule". At the other extreme, Murphy J was clearly prepared to abandon it as stating "no satisfactory general principle" .

In searching for an alternative formula to give guidance beyond the Delphic suggestion that recovery was permissible in "special" or "exceptional cases" , the Justices in Caltex found different pointers on the compass which would point them away from the rocks of "indeterminate liability" to an "unascertained class" . Clearly, that was regarded as unacceptable as a matter of policy of the common law, out of harmony with past legal authority and prone to cause serious economic outcomes.

For Gibbs J in Caltex , the relevant principle was the knowledge of the defendant (or means of knowledge) that the plaintiff individually would be likely to suffer economic loss as a consequence of the negligence. Mason J's view was similar , although the Privy Council rightly observed that Mason J's citation of Canadian authority suggested acceptance of a principle somewhat broader than that adopted by Gibbs J. Stephen J , on the other hand, listed five considerations as pointing to that "close degree of proximity between the defendant's conduct ... and the economic loss ... suffered [by the plaintiff]" which would attract the legal duty of care notwithstanding the fact that the damage suffered was purely economic. Foresight of the impact on an identifiable victim was one, but not even the main, consideration for Stephen J's approach. Jacobs J, whilst commencing his reasoning from the exclusionary rule, considered that the case in hand fell within an exception or re-exposition of the rule because of the demonstration of the "physical propinquity" between the damage done to the third party's property and consequential physical effects on the property interests of the plaintiff . Murphy J considered that the "control mechanism" for which all members of the Court were searching was to be found in policy. This he expressed as: "Persons causing damage by breach of duty should be liable for all the loss unless there are acceptable reasons of public policy for limiting recovery."

It emerges from this analysis of Caltex that, having rejected adherence to the exclusionary rule, no member of this Court considered that the reasonable foreseeability of harm to the plaintiff's economic interests was sufficient as a "control mechanism" to limit the occasions of liability to a manageable and acceptable scope . Foreseeability of harm would not adequately confine legal liability to a determinate class. Hence the attempt in the reasoning of the several Justices to add a notion of proximity, either explicitly or implicitly by the requirement that the defendant should have knowledge, or means of knowledge, of the risk of damage to the interests of the plaintiff individually. Except in the reasons of Murphy J , there was less explicit recognition of the importance of a policy evaluation to the final decision. Yet it will be observed that already, in the several reasons given in Caltex , the three elements later identified in Caparo were forming: foreseeability, proximity and policy. This is the way that these developments usually happen in the common law. Out of a multitude of cases, and of judicial decisions in those cases, comes a new legal concept - eventually emerging fully formed from the efforts of the past.

In later decisions of this Court, in cases where liability for pure economic loss was claimed, the acknowledgment of the place of policy that has come to characterise this area of the law (and to which Callinan J correctly refers in his reasons ) entered with growing candour into this Court's discourse. Thus in Bryan , the joint judgment approached the issues recognising that, in this context (as distinct from "more settled areas" concerned with ordinary physical injury to the person or property of the plaintiff), reasonable foreseeability of injury or damage to the plaintiff would not suffice to found a duty of care and thus legal liability. To demonstrate such a duty of care in a case framed in negligence, it was necessary to show that "there exists a relationship of proximity between the parties with respect to both the relevant class of act or omission and the relevant kind of damage". In cases of alleged liability for pure economic loss, the "notion of proximity ... is of vital importance" . But, in Bryan , their Honours went on to refer to "policy considerations". Two were expressly singled out as critical to the policy analysis. These were the avoidance of the imposition of indeterminate liability and of liability for loss which was the ordinary outcome of economic competition in the marketplace. Such considerations were not put forward in Bryan as separate criteria or "principles". Nowhere was it suggested that they constituted distinct, identifiable and particular preconditions to the existence of liability in negligence. They were subsumed, correctly in my view, amongst the policy considerations to be weighed once the plaintiff had established foreseeability and proximity.

In Hill v Van Erp the notion that proximity could be used as a universal concept to encompass, in every case, the prerequisite of reasonable foreseeability and to answer the evaluation of policy considerations was rejected. Dawson J explained why this was so: "[I]f considerations of policy underlie and enlighten the concept of proximity, and if nearness and closeness are neither sufficient nor necessary to establish a relationship of proximity in all cases, then it cannot be said that any unifying common element has emerged which can adequately be described as proximity."

In so far as proximity was being pressed into meaning more than "nearness and closeness" or "neighbourhood" in the legal sense, it was being asked to fulfil a purpose which it was not historically designed to perform. Proximity was incapable of fulfilling, unaided, the function of demonstrating the existence or absence of a duty of care. If, on the other hand, proximity were to be confined to its original historical purpose as a measure of "nearness and closeness" between the parties in dispute, it could yet provide a meaningful gateway, in addition to reasonable foreseeability of harm, to afford the starting point for the allocation of a legal duty of care or exemption from its burden. Then, it would remain necessary (as the Court came to do in Hill v Van Erp ), to weigh candidly the competing policy considerations relevant to the imposition of a legal duty of care.

As I read the recent opinions of this Court, illuminated now by parallel developments in all of the common law jurisdictions which I have mentioned (with the possible exception of the United States), and as I view the successive rejections of "reasonable foreseeability", "reliance" and "proximity" as universal determinants of the existence or absence of a legal duty of care, I am reinforced in my opinion about the way cases of this kind must be decided . A three-fold approach or methodology of reasoning is required. Even if foreseeability and proximity are first established, there is no escaping the evaluation of the competing arguments of policy for and against the attachment of legal liability in the particular case. It cannot be escaped by formulating so-called new "principles" of universal operation. To do this would be to show that we have learnt nothing from the successive rejections of earlier judicial attempts to express the criteria for the existence of a duty of care in terms of universally applicable rules devoid of policy choices.

As against the approach which I favour, it has been said that the three identified elements are mere "labels" . So indeed they are. But, equally, negligence itself is a "label". So are the criteria or "principles" of determinacy of liability, vulnerability to risk, autonomy of the individual, economic competitiveness and predictability of outcome. Labels are commonly used by lawyers. They help steer the mind through the task in hand. The alternative, where no single accepted touchstone for legal rights and duties exists and where precise rules (even if called "principles") are rejected as unacceptable, is an unsatisfactory resort to nothing more than tautologous generalisations such as the holding that a case is "special" or "exceptional".

It may appear that the third of the Caparo considerations ("fair, just and reasonable") partakes of the last-mentioned defect. It does not. The third consideration is simply another way of directing the mind to the questions of legal policy which need to be weighed in reaching a final conclusion in the particular case. It permits consideration of the many questions identified in past cases as relevant to attaching or denying a duty of care . They are not "rules" in themselves. They are not "principles". They do not rise to a level of such generality. But they are considerations which, depending on the particular facts, need to be evaluated, if the other preconditions (foreseeability and proximity) are established, before the final decision is made to uphold or reject the existence of a legal duty of care.

Critics have complained that the three-fold approach is not helpful. If the courts have recognised as "control mechanisms" the avoidance of indeterminate liability and non-interference with ordinary business conduct, it is asked, why not say so directly ? Why not simply adopt these considerations as legal rules?

The answer to this question depends upon the importance attached to returning legal liability for pure economic loss to a conceptual framework common to all other actions expressed in terms of the tort of negligence. It also depends upon the level of abstraction in which such concepts are to be stated for legal purposes. Considerations of indeterminacy of liability, vulnerability to risk, the autonomy of the individual and market competitiveness are not issues relevant to all negligence actions. It is therefore quite inappropriate to elevate them so that they are legal preconditions to the existence of a duty of care in negligence or "principles" to be applied in deciding whether the duty exists in the particular case. They are not even essential or relevant to every case framed in negligence where the damage claimed is purely of an economic character, without physical injury to the plaintiff's property or person. What is therefore needed is a more general or conceptual methodology or approach which provides the heading to which these considerations may be assigned when, in the particular case, they are considered relevant. Conceptually, they all belong to the evaluation of considerations of policy. Each of them is a powerful negative policy reason for inhibiting the extension of legal liability in negligence for pure economic loss.

Finally, it is said that the three-fold approach imposes upon lawyers advising clients, trial judges and even appellate courts unreasonable burdens and upon clients intolerable uncertainties and costs. The answer to this complaint is that such burdens and uncertainties already exist. They do not disappear simply because the policy considerations are artificially squeezed into the "labels" of "foreseeability", "reliance" or "proximity". They certainly do not disappear because new rules or "principles" are invented in the fond hope, once again, of establishing universal norms. The policy evaluation stubbornly remains. As it seems to me, it is better by far that it be addressed with candour and recognised for the legal function that it performs. I do not accept that such a function is beyond the evaluative powers of judges and lawyers. Indeed, they already perform that evaluation. Caparo and its methodology simply bring the judicial choices into the open.

In this area of the law, there has been a candour about the influence of legal policy which surprises some commentators, and even some judges, used to the pretence that the law provides for every problem a simple rule yielding an unarguable outcome. In this field of negligence law, at least, it does not. This is not the time to return to pretended classifications, narrow "exceptions", newly invented "principles" or rules derived from the apparent justice of a hard case, or unconceptual categories. The law of negligence in cases of claims for pure economic loss is completely unsatisfactory. This Court's duty is not to search for more "rules" which will last a moment but fail to afford more than fleeting and particular guidance as new and different circumstances present themselves for decision. Our duty is to afford an approach or methodology which is universal to the tort of negligence and appropriate to the particular sub-category of negligence where the breach in question has occasioned pure economic loss. Only the Caparo formulation achieves that goal. We should adopt it. In doing so we will be taking no bold step. The methodology or its variants are, as I have shown, already applied in England, New Zealand and Canada - the countries of the common law whose legal systems most closely approximate that of Australia. Striking out with our own new "principles" is in my view quite the wrong way to go. It is doomed to fail. Application of the three-stage approach

Foreseeability: When the problem in the present appeal is approached by the methodology which I favour, the answer to the first question (foreseeability) is readily given. It is beyond serious argument in this case that damage to the Perre interests of the type that occurred was, or ought reasonably to have been, foreseen by Apand . There was a real risk of its occurring. It was not a risk which was fantastic or far-fetched . In its application to breach of duty, the test of foreseeability has been described as "undemanding" . When considering the existence of a legal duty of care a more generalised inquiry is required . However, when that inquiry is conducted in the present matter, it leads to only one conclusion. Not only ought Apand to have foreseen damage of the kind that might occur to the Perre interests; the evidence discloses that it actually did foresee precisely the kind of damage that eventuated, even if it did not turn its attention to the particular individuals who would be afflicted with that damage if it occurred.

The reasons of other members of this Court set out the letter written to Apand in April 1990 by the Victorian Department of Agriculture and Rural Affairs and the circular warning which Apand's supply manager sent soon afterwards to senior officers of the company, including those in charge of operations in South Australia. These letters specifically warned of the dangers of spreading bacterial wilt by planting seed which had not been inspected and certified by the Victorian Department. Expressly, the latter drew attention to the Western Australian prohibition for five years on the entry of potatoes grown within a 20 kilometre radius of an outbreak of bacterial wilt. This message was reinforced in June 1990 by the further letter which Apand's national potato supply manager wrote to the managers in four States, including South Australia. Notwithstanding these warnings, Apand pressed ahead with the use of its non-certified seed. The damage to the Perre interests which ensued was precisely that which was foreseen. The first consideration is therefore satisfied.

However, that does not entitle the Perre interests to recover damages in negligence. To attach liability to this consideration, without more, and to hold it sufficient to give rise to a legal duty of care on the part of Apand to the Perre interests, would impose "so vast a restriction on individual freedom of action" that more is necessary to warrant the imposition of a duty of care which the law will enforce . Foreseeability could not be the sole determining factor because that task is "beyond its capacity" . As Professor Fleming explained long ago : "Duty presupposes foreseeability of damage but foreseeability does not entail a necessary inference of duty."

Proximity: The second step therefore requires consideration of the relationship between the parties, in the sense of the relationship between the relevant classes of act or omission on the part of Apand and the relevant kinds of damage suffered by the Perre interests. Can a relationship of this type between them be characterised as one of "proximity" or "neighbourhood" in the relevant legal sense? The Full Court thought not . Their Honours were affected by the fact that there was no relationship "equivalent to contract" between Apand or the Sparnons (on the one hand) and the Perre interests (on the other). Nor were the latter in any sense the object of the arrangements between Apand and the Sparnons. The Perre interests stood neither to gain nor to lose from the initiatives which Apand took in their vicinity. By inference, they just happened to be in the wrong place at the wrong time.

With respect, I cannot accept this analysis. Whether or not the particular conversation between Mr Hughes for Apand and Mr Frank Perre about the Perres becoming one of Apand's contract growers ought to have alerted Apand to the existence and identity of the Perre interests and the particular risks they ran, matters not. I shall assume that the Full Court was right to conclude, on the basis of the primary judge's findings, that the information was "casually given" and not such as to establish specific notice of the individual risk to the Perre interests explicitly identified as such to Apand . But that is by no means determinative of the consideration of "proximity" in the circumstances of this case.

Apand through its employees, including Mr Hughes, knew or ought to have known of the existence of other potato growers within very close proximity to the Sparnons' property into which it introduced the uncertified seed. Having received the letters of warning from Apand's central office, the senior officers of Apand, including in South Australia, would have found it simple to discover (if they did not already know) the identity of each and every potato farmer within the critical 20 kilometre distance of the Sparnons' property. Similarly, Apand could readily have identified any farmers who "harvested, cleaned, washed, graded or packed with equipment or in premises with or in which" potatoes grown within 20 kilometres of such an outbreak would have been handled.

As a matter of fact, the evidence showed that each of the Perre interests carried on their respective activities within extremely close physical proximity to the Sparnons' property. Thus, Warruga Farms Pty Ltd was, at its closest point, only about three kilometres north of that property . Perre's Vineyards Pty Ltd carried on its business on land directly opposite the Warruga Farms property . Prior to the outbreak, Perre's Vineyards leased its premises to Warruga Farms for potato washing and packing. The twelfth to fifteenth appellants had separate properties adjoining the southern boundary of the Warruga Farms property. The southern-most of these was only about two kilometres north of the Sparnons' property. Those individuals entered into the Rangara Joint Venture to grow potatoes on the properties and to sell them to Warruga Farms for delivery to Perre's Vineyards for washing and packing and thence, in large proportion, for export to Western Australia .

Although it is true that the evidence, as found, did not establish actual physical damage to the property of the Perre interests, that consideration as well as the identification of risk to a known and identified party can now be seen as "control devices" which the law adopted in its attempt to limit the ambit of liability and to avoid its extension to an indeterminate class. There was no risk of indeterminacy in this case. The ambit of the reasonably foreseeable, indeed known, vulnerability was measured by precise considerations of geographical proximity. The outer boundary was relevantly circumscribed by potato growing interests within a radius of 20 kilometres from a farm to which the non-certified seed was sent for planting. This was a risk which Apand took with its eyes open. Potato farmers within that radius were in a relationship of proximity or neighbourhood because of their vulnerability arising from an almost contiguous physical propinquity to a farm to which Apand decided to introduce the uncertified seed as part of its winter crop experiment. The second consideration is therefore also satisfied.

Policy: When I turn to the third consideration, it is necessary to weigh the two particular policy reasons which, in the past (and, it was suggested, in this case), have emerged to justify a rule excluding a legal duty of care to a plaintiff who has suffered no physical damage to its person or property but only pure economic loss.

I refer, first, to the concern to avoid imposing legal liability upon an indeterminate class for indeterminate amounts . This is a legitimate and justifiable concern. It would, for example, allow the law to draw a rational line which would deny recovery to the local store in the Perres' town whose income had dropped because of the loss to the Perre interests (and hence to the income of its principals and employees) following the loss of the export trade to Western Australia. It would permit exclusion of a legal duty of care to a trucking firm which, before the blight, carried the potatoes to Western Australia. It would exclude consumers in that State, forced to purchase potatoes, or the crisps manufactured from them, from other more distant and expensive suppliers. A line must be drawn to limit indeterminate liability. What permits it to be drawn in this case, and not to exclude all of the Perre interests, is the specific foresight of potential damage to potato producers in the position of the Perres, operating in such close proximity to the Sparnons. In this sense the Perres were particularly vulnerable to the conduct of Apand, as it ought to have recognised and in a general sense did.

There was a high likelihood, which would have been evident to Apand, that potato producers in South Australia would ordinarily seek to take advantage of their geographical proximity to Western Australia to export to the lucrative market in that State any potatoes which they did not sell to Apand itself. Apand knew, and had recognised, the risk of bacterial wilt in using uncertified seed, given the terms of the Western Australian regulation. It is that regulation which answers the complaint about indeterminate liability. The principal policy reason for the operation of the former exclusionary rule is therefore inapplicable in this case. The regulation imposed a form of quarantine which was, relevantly, specific to potato growers affected by what Apand, with its access to expert knowledge and advice, did. The regulation had this effect as much on the Perre interests as on the Sparnons. To differentiate between their several entitlements to succeed in an action at law in negligence against Apand on the ground that a duty of care was owed to the Sparnons but not to the Perre interests, is artificial and unjust. Such differentiation on no basis better than that there was no direct damage to the Perre interests' properties should be rejected. As a matter of policy it can be rejected in this case because the concern about indeterminate liability is inapplicable.

Nor is this a case where to hold Apand to a legal duty of care would be to interfere unreasonably with its economic freedom, its autonomy and the competitive operation of the marketplace . As a matter of policy, the law will generally uphold the right of a party lawfully to gain profit although doing so will occasion economic loss to others. That is often the very way in which the market operates in our form of economy. Apand certainly had an economic interest in disposing of the non-certified seed when its original Saturna experiment was called off. But Apand was not entitled to pursue that interest in a way that constituted a breach of the law.

Apart from any other reasons which may apply in this case, as Hayne J has pointed out, the fruit and plant quarantine laws of South Australia absolutely prohibited the importation or introduction into the State of South Australia of any plant (defined broadly enough to include seed) affected by bacterial wilt . Yet upon the premises established by the factual findings not now in dispute, this is precisely what Apand did. It is not necessary to consider whether, absent such statutory prohibitions, illegality otherwise affecting what Apand did in relation to the Perre interests would have been sufficient in any case to put a limit on Apand's economic freedom so far as it impinged on the legitimate interests, economic and otherwise, of the Perre interests. No reason of policy arising from the ordinary deference which the law pays to that economic freedom and market competitiveness prevents a finding that a legal duty of care existed in this case.

I do not consider that any other reason of legal policy, derived by analogy from earlier decisions in claims for pure economic loss (or from the so-called "exceptions" to the exclusionary rule), warrants a denial of the existence of a legal duty to which the foregoing considerations strongly point. So I now look back and check this conclusion with the general objects of the law of negligence in mind. The conclusion is confirmed. The basic question that lies at the heart of the tort is asked: ought Apand to be under a legal obligation to observe care to prevent damage to the Perre interests from the risk inherent in its activities, the nature of which Apand actually foresaw or which was reasonably foreseeable to it ? The answer which I would give to that question is in the affirmative. Apand owed the Perre interests a duty of care. An action based on the tort of negligence lies to enforce that duty for a proved breach which results in damage. It is not excluded merely because the damage suffered by the Perre interests was purely economic loss. Differentiation of the Perre interests

McHugh J and Hayne J have severally concluded that, although a duty of care was owed by Apand to some of the Perre interests, the law differentiates between the appellants so that no duty of care was owed to others. The other members of the Court, in their respective reasons, make no such differentiation. They consider that the duty of care was owed to each of the Perre interests but that the quantification of their respective damages remains to be determined at trial. I agree with the majority view.

Taking as an example the differentiation of the position of Warruga Farms, suggested by Hayne J, I consider, with respect, that it is unjustified. It is true that it was Warruga Farms that actually exported the potatoes to Western Australia. The evidence, although in some respects unsatisfactory, demonstrates the closely integrated nature of the operations of the several Perre interests in the growing and processing of potatoes. Not only were their respective properties physically proximate (abutting or facing as each did the Warruga Farms property), they were all inextricably interconnected with the Warruga Farms business. They were not in the position of the town store, contract carrier, or purchaser or consumer down the line of supply in Western Australia. Damage to Warruga Farms inevitably and immediately caused damage to Perre's Vineyards, the individual Perre family members and Rangara.

In negligence law, lines must be drawn. But it would be highly artificial to divide some of the members of this integrated commercial operation from others. Warruga Farms would not have been able to put their potatoes into the Western Australian market (attracting the force of the regulation) without the supply of raw materials and of the services of washing and grading which the other Perre interests provided. The differentiation in the respective entitlements of each of those interests, and the avoidance of double counting, undoubtedly present practical problems at the stage at which the damages are quantified. But that stage has not yet been reached in the course which the trial took. The several claims should now be returned to trial. It is there that the damages lawfully recoverable by each of the appellants may be calculated according to law. In finding otherwise, the Full Court erred. Orders

The orders proposed by Gaudron J should be made.

HAYNE J. In 1991, the respondent sold seed potatoes to D C, S K & M D Sparnon, a firm of commercial potato growers in the Riverland district of South Australia. The seed potatoes produced a crop that in April 1992 was found to be suffering from bacterial wilt, a potato disease. Western Australian regulations permit entry of potatoes into that State only if certified that two conditions were met: first, that they were not grown on a property situated within 20 kilometres from a known outbreak of bacterial wilt that had occurred within the last five years; secondly, that they were not (unless otherwise approved) harvested, cleaned, washed, graded or packed with equipment, or on premises, with or in which potatoes grown within 20 kilometres of a known outbreak of bacterial wilt, detected within the last five years, have been handled.

In 1992, the appellants grew and packed potatoes on properties between about 2 and 3½ kilometres from the Sparnons' property where the diseased crop was growing. It will be necessary to describe the role of each of the appellants more carefully but for the moment it is enough to say that much of the business of each of them was connected to the production of potatoes for sale in Western Australia. The disease did not spread to the appellants' properties. But because of the outbreak of bacterial wilt, they could no longer process potatoes for export from South Australia to Western Australia. The appellants contend that as a result of being denied this market for their potatoes they lost money. They, and the purchasers of the seed potatoes (the Sparnons), sued the respondent and others in the Federal Court of Australia. Both at first instance and on appeal to the Full Court of the Federal Court the Sparnons succeeded in their claim against the respondent but the appellants failed. The appellants now appeal to this Court. The respondent no longer challenges the judgment obtained by the Sparnons. The central question raised on the appeal to this Court is whether the respondent owed the appellants (or any of them) a duty of care when the losses allegedly suffered by the appellants are wholly financial and include no element of damage to person or tangible property. The background

In 1992, the respondent commanded about 60 per cent of the potato crisping industry in Australia. It bought large quantities of potatoes from growers in the Eastern States and South Australia. It had a large potato storage at Tynong in Victoria and a modern factory at Regency Park, South Australia where it made potato crisps. The respondent does not carry on this business now. In January 1993, after the events that gave rise to this litigation, it sold its potato crisping business.

The respondent carried out research into and development of new varieties of potato, including potatoes that could be grown as a winter crop. In 1987, the respondent arranged for the importation into Australia of tissue culture for the Saturna variety of potato. It was thought that this variety might be a suitable winter crop. The respondent set about developing production of Saturna seed through the Victorian Seed Potato Certification Scheme - a scheme operated by the Victorian Department of Agriculture for the multiplication of seed potatoes that will, as far as possible, be disease free. (There is, and can be, no absolute assurance against transmission of disease by seed potatoes.) By March 1990, 6 tonnes of third generation seed had been produced. Five of the 6 tonnes was delivered to the respondent.

In the middle of 1990 the respondent decided not to go ahead with the development of Saturna potatoes; the tubers were thought not to be the right size and shape for regular use in its business. Accordingly, it was decided to grow the 5 tonnes of seed and process the resulting crop.

The Saturna seed was grown at Garfield, Victoria (in the Koo Wee Rup Swamp area) by GMT Trading Pty Ltd, a grower whose principal was a Mr G P J Tymensen. (Mr Tymensen was held in high regard by the respondent and others in the Victorian potato industry.) The Koo Wee Rup Swamp area has long been used for potato growing. It is, however, low lying and susceptible to aphids and the diseases they transmit, and it has seen outbreaks of other potato diseases and pests. It is an area not used in the Seed Potato Certification Scheme and Mr Tymensen had never been a grower of seed for that scheme.

After the Saturna seed had been planted at the Tymensen property, but before harvest, one of the respondent's employees asked whether the Saturna variety might be used in Queensland. It was suggested that the Tymensen crop be used as a seed crop but one of the respondent's field officers (Mr Moorthy) expressed some reservations about using the crop in this way; nevertheless he told others at the meeting at which these discussions occurred that he would "look at it". Those at the meeting (which was held at Pakenham in Victoria, and became known at the trial as "the Pakenham Meeting") agreed that Mr Tymensen's crop could be used as seed to enable multiplication of the Saturna variety.

In 1991, Mr Tymensen produced about 86 tonnes of Saturna potatoes. He kept 10 tonnes and delivered the rest to the respondent. The respondent processed some of these potatoes but distributed the rest to 13 growers in Queensland, South Australia, Victoria and New South Wales for further trial and multiplication. In June 1991, the respondent sold 2 tonnes of the Saturna seed to a grower in South Australia (Virgara Bros of Virginia). Some of this was planted and 17 tonnes were harvested in December 1991.

In December 1991, a representative of the respondent (Mr John Hughes) invited five of the respondent's contract growers of potatoes in South Australia to grow an experimental crop of Saturna potatoes. One of these growers was the Sparnon partnership and the Sparnons agreed to participate. Accordingly, in December 1991 the respondent made the sale to the Sparnons that I referred to at the start of these reasons. The outbreak of bacterial wilt

Five of the six experimental crops of Saturna potatoes, planted in South Australia in 1992 at the respondent's initiative, were found to be affected by bacterial wilt. (Four of the five growers approached by Mr Hughes grew crops that were affected; a crop grown by Virgara Bros was also affected.) When the outbreaks of the disease were discovered, representatives of the respondent looked closely at the paddock where Mr Tymensen had grown his first Saturna crop. The trial judge (von Doussa J) found that they discovered regrowth potatoes that showed clear evidence of bacterial wilt. The affected potatoes were in an area of about 4 to 5 square metres in one corner of the paddock. Although the respondent denied that Mr Tymensen's land was the immediate source of the infection of the crops grown in South Australia, the trial judge found that it was. He also found that harvesting and grading the potatoes from this small area could have caused the disease to be distributed widely through other tubers harvested and graded at the same time or by use of the same machinery. The respondent did not dispute these findings on the hearing of the appeals to the Full Court of the Federal Court .

How Mr Tymensen's crop was infected with bacterial wilt was a matter of considerable dispute at trial. The trial judge found that in October 1990 several properties in the Koo Wee Rup Swamp area were flooded. Mr Tymensen's property was one of those affected - some of his land, including the corner where bacterial wilt was later discovered, being submerged for a day or two. Bacterial wilt had previously been discovered (in 1986 and 1990) on the property of a Mr Summers. That property abutted the land of a Mr O'Sullivan which, in turn, abutted Mr Tymensen's land. The trial judge found that: "as a matter of probability the likely explanation for bacterial wilt in Mr Tymensen's first Saturna crop is that the flood in some manner carried it there from Mr Summers' property." In the Full Court, the respondent contended that this finding was not open to the trial judge but the Full Court rejected that contention . The Sparnons' claim

Although not challenged on the appeal to this Court, it is convenient to notice now the basis on which the respondent was held liable to the Sparnons. As is apparent from what I have said earlier, the trial judge found that the respondent sold the Saturna seed potatoes to the Sparnons. That is, as the Full Court put it in its reasons for judgment, "the arrangement made ... in about December 1991 for the Sparnons to grow an experimental winter crop with two tonnes of Saturna potatoes was a contractual one." The trial judge found that warranties of fitness for purpose and merchantable quality were implied in that contract pursuant to ss 14 I and 14 II of the Sale of Goods Act 1895 (SA). The respondent was found to have breached both of these warranties and to be liable to the Sparnons for damages for breach of contract.

The trial judge also found that the respondent owed the Sparnons a duty of care, that it breached that duty, and that the Sparnons were entitled to recover damages for this negligence. His Honour said that: "... the duty imposed by the relationship was to take all reasonable steps to ensure that seeds which Apand provided to its growers had not come from a source where there was a real risk which Apand knew about or should have foreseen that the seeds might be infected by pests or disease. A real risk is one that people well versed in the nature of potato pests and diseases, such as the Apand officers, would not brush aside as far fetched ." The finding made of breach of that duty was expressed by the trial judge as follows: "I think with the benefit of the knowledge collectively possessed by the officers at the Pakenham meeting, they should have foreseen the risk that the flood may have spread bacterial wilt to the area of Mr Tymensen's property where the crop was growing, and should have foreseen the risk that the crop could be infected with low levels of bacterial wilt that may not be seen in the crop on the inspections which had occurred and were planned to occur." Although this finding was made in the context of considering the claim made by the Sparnons, it is a finding that is significant in considering the present appeal. The appellants' claims at trial

It is convenient to say something at this point about the different interests of the various appellants and the claims that they made. On the hearing of the appeal, they were divided into four: the natural persons, the Rangara Joint Venture, Warruga Farms Pty Ltd and Perre's Vineyards Pty Ltd. The natural persons (six named couples) owned two pieces of land - the Warruga land (owned by Frank and Caterina Perre, Pasquale and Mary Perre, Guiseppe and Maria Perre and Jim and Frances Perre) and the Rangara land (owned by Pasquale and Grace Perre and Francesco and Maria Perre). Warruga Farms Pty Ltd was a company that grew potatoes on the Warruga land. (The shareholders in Warruga Farms Pty Ltd were Pasquale and Mary Perre, Guiseppe and Maria Perre, Jim and Frances Perre and Perre's Vineyards Pty Ltd.) Warruga Farms processed the potatoes it grew (and other potatoes it bought from the Rangara Joint Venture) and between 1990 and May 1992 it had sold an increasingly large part of its crop to purchasers in Western Australia.

The Rangara Joint Venture was a venture between Pasquale and Grace Perre and Rangara Pty Ltd (the shareholders in which were Francesco and Maria Perre). The joint venturers grew potatoes on the Rangara land and sold the potatoes to Warruga Farms.

Perre's Vineyards Pty Ltd (the shareholders in which were Frank and Caterina Perre) owned the potato processing facility and the land on which it stood and let the facility to Warruga Farms on a tenancy at will.

Warruga Farms alleged that it had suffered loss because once the outbreak of bacterial wilt was discovered on the Sparnons' property it could not, thereafter, earn income by exporting potatoes to Western Australia. (It also alleged that it had incurred various expenses in attempting to mitigate that loss.) The Rangara Joint Venture alleged that it lost income as a result of losing the sales it would have made to Warruga Farms but for the inability of Warruga Farms to export to Western Australia after the outbreak was discovered. Perre's Vineyards alleged that as a result of these events it lost its tenant and could not use its processing facility. The owners of the Warruga land alleged that the value of their land had decreased. The owners of the Rangara land alleged that they had sold their land at a loss. None of these allegations about the appellants' losses was examined at trial, the trial judge having ordered that issues of liability be tried separately before issues of quantum of damage. For present purposes, what is of particular significance is that in each case the loss allegedly suffered is a loss said to be caused by the growers and processor being prevented from directly or indirectly selling potatoes they had grown or processed into the Western Australian market. And the loss was said to be suffered because of the loss of future sales. No allegation was made that the outbreak of bacterial wilt (and consequential regulatory prohibition on import into Western Australia) prevented any appellant from performing any existing contract for the sale, supply or processing of potatoes. Perre's Vineyards complained that Warruga Farms had brought its tenancy at will of the processing facility to an end but it did not allege that this termination constituted a breach of its contract with Warruga Farms. Trial and appeal to the Full Federal Court

The trial judge held that the respondent owed no duty of care to any of the present appellants and entered judgment for the respondent. The Full Court dismissed the appeal against that judgment. It said : "At the material time, [the respondent] had neither knowledge nor means of knowledge that the Perre interests individually, rather than as a member of an unascertained class, would be likely to suffer economic loss as a consequence of its negligence in supplying infected or potentially infected seed to the Sparnons. Even if one were to expand the permissible scope of knowledge or means of knowledge to a defined and small class of persons capable of ascertainment, we do not think the authorities presently indicate that that would be, of itself, sufficient, unless there were some other particular circumstances which drew to [the respondent's] attention that it should, in taking the steps which it was found negligently to have taken qua the Sparnons, have also had regard to the interests of the members of that particular defined and small class of persons. There is no particular consideration of that nature in the present circumstances." Standing alone, this passage from the reasons of the Full Court might be taken as suggesting that the respondent was entirely unaware of what might happen if it were to do something that could or would lead to an outbreak of bacterial wilt. In fact, however, evidence at the trial showed that the respondent was well aware how bacterial wilt was spread, what would happen to growers if there were an outbreak near their properties, and what kinds of criticism might be levelled against the respondent if it were associated with an outbreak. The respondent's knowledge

In June 1990, the National Potato Supply Manager of the respondent sent a memorandum to the respondent's Manufacturing Managers in Queensland, Western Australia, New South Wales and South Australia. In part, that memorandum read: "The Potato Industry in general is concerned at the increasing rate of incidence of Bacterial Wilt and Potato Cyst Nematode in the past few years. Bacterial Wilt and Potato Cyst Nematode, both of which are 'notifiable diseases' are largely controlled by strict quarantine and long rotations between potato crops ie No potatoes grown for 5 to 10 years. The economic impact on a grower who has the disease on his farm can be disastrous - he is unable to continue to grow potatoes in that part of the farm and unable to sell his infected farm . ... We have had contracts for supply with growers who have these diseases, in most states. Recent outbreaks in Qld of Bacterial Wilt, continues the trend. We can anticipate that should it be believed by growers that CCA was instrumental in the spread of the disease to their farm that litigation would be considered. CCA operations must implement adequate controls. Growers still complain of 'old' potatoes being returned in supposedly 'clean' bins. By way of example the Department of Agriculture and Field Officers in the USA where these diseases exist use disposable plastic overshoes etc. The Western Australia Dept of Ag will not allow potatoes sourced within 20 kms of a known outbreak of Bacterial Wilt, into their state. The major cause of spread is through growers buying non certified seed. This often occurs where crops are grown out of season ie Winter harvest and regular certified seed is unavailable at that time of year. Hence there are major outbreaks throughout the Riverina of NSW, Maitland and now Southern Queensland. Whilst seed is the major cause another is Potato Bins. It is in our own interest to maintain vigilance to protect our sources of supply." (Emphasis added) That is, less than two years before the outbreak of bacterial wilt on the Sparnons' property, a senior employee of the respondent warned other senior staff to "implement adequate controls" against the spread of bacterial wilt because, if the disease occurred, "[t]he economic impact on a grower ... can be disastrous" and because "[t]he Western Australia Dept of Ag will not allow potatoes sourced within 20 kms of a known outbreak of Bacterial Wilt, into their state." He also warned that "[t]he major cause of spread is through growers buying non certified seed." Thus, within two years of the events of which the appellants now complain, a senior employee of the respondent warned of the very consequences that would follow for these appellants if bacterial wilt were spread through a neighbouring grower buying non-certified seed, as the Sparnons did from the respondent.

When the respondent sold the Saturna seed to the Sparnons, it knew that there were other potato growers operating near the Sparnons' property. At the trial the appellants tried to prove that, at the time of the sale to the Sparnons, the respondent had known of their particular activities. That attempt failed. Accordingly, the most that could be said is that the respondent knew that there were other potato growers in the area and knew that it was probable that those growers processed their crops nearby. The nature of the loss

The appellants' loss was wholly economic. There was no demonstrated escape of the disease to their land or to their crops.

A person injured by the negligence of another will very often suffer adverse economic effects - loss of wages, the cost of repairing damaged property and the cost of obtaining medical treatment. And monetary damages are the only salve the law offers to a person affected by the negligent act or omission of another. Yet the law has long treated "pure economic loss" differently from other forms of injury. Although a person is just as much injured if the value of some intangible is diminished as that person is if bodily injury or injury to tangible property is sustained, separate and difficult questions are seen as arising in the first case. Especially is this so when the loss that is suffered is a loss in the course of carrying on a business. Making a loss of that kind is often no more than one of the ordinary consequences of participation in a market economy. The market economy treats rivalry between participants as an essential and defining feature: rivalry in which each participant seeks to maximise its profit and market share at the expense of all other participants in that market.

Two threads, then, can be seen as important in the development of the principles governing liability for negligently caused pure economic loss. First is the desire to avoid "liability in an indeterminate amount for an indeterminate time to an indeterminate class" . Second is the concern not to establish a rule that will render "ordinary" business conduct tortious. Deciding whether one person owed a duty of care not to cause pure economic loss to another requires consideration of both these matters. Foreseeability of injury is essential but not enough. But that test (as it has developed) often extends widely. The decisions in the area of pure economic loss reflect the search for some further mechanism of controlling liability (further, in the sense of additional to foreseeability of injury) . In recent years there has been much discussion of whether "proximity" is that mechanism. Proximity

Often (but not always) a negligent act or omission will cause physical consequences in a limited area or to a limited group of people - those who can be described as neighbours or proximate. Lord Atkin said of his famous statement of the "neighbour" principle that: "This appears to me to be the doctrine of Heaven v Pender , as laid down by Lord Esher (then Brett MR) when it is limited by the notion of proximity introduced by Lord Esher himself and A L Smith LJ in Le Lievre v Gould . Lord Esher says: 'That case established that, under certain circumstances, one man may owe a duty to another, even though there is no contract between them. If one man is near to another, or is near to the property of another, a duty lies upon him not to do that which may cause a personal injury to that other, or may injure his property.' So A L Smith LJ: 'The decision of Heaven v Pender was founded upon the principle, that a duty to take due care did arise when the person or property of one was in such proximity to the person or property of another that, if due care was not taken, damage might be done by the one to the other.' I think that this sufficiently states the truth if proximity be not confined to mere physical proximity, but be used, as I think it was intended, to extend to such close and direct relations that the act complained of directly affects a person whom the person alleged to be bound to take care would know would be directly affected by his careless act." But, as the cases decided since Hedley Byrne & Co Ltd v Heller & Partners Ltd and Mutual Life & Citizens' Assurance Co Ltd v Evatt show, the search for a criterion or criteria that will identify a relationship between plaintiff and defendant as one that is sufficiently "proximate" to hold the defendant liable for pure economic loss sustained by the plaintiff because of the defendant's negligence is attended by great difficulty. And although the "relationship of proximity" may be a useful description of the result of the decision whether, in particular circumstances, the defendant owed a duty to the plaintiff not to cause pure economic loss, it is only in that sense that the relationship of proximity "remains the general conceptual determinant and the unifying theme of the categories of case in which the common law of negligence recognises the existence of a duty to take reasonable care to avoid a reasonably foreseeable risk of injury to another" . As five members of the Court said in Burnie Port Authority v General Jones Pty Ltd : "It is true that the requirement of proximity was neither formulated by Lord Atkin nor propounded and developed in cases in this Court as a logical definition or complete criterion which could be directly applied as part of a syllogism of formal logic to the particular circumstances of a particular case . As a general conception deduced from decided cases, its practical utility lies essentially in understanding and identifying the categories of case in which a duty of care arises under the common law of negligence rather than as a test for determining whether the circumstances of a particular case bring it within such a category, either established or developing ." Statements to similar effect are to be found in Hill v Van Erp . To search, in these circumstances, for a single unifying principle lying behind what is described as a relationship of proximity is, then, to search for something that is not to be found.

And it is for like reasons that the problem of identifying when a duty of care to prevent economic loss should be found to exist cannot be solved by discarding references to proximity and substituting a test that is (or includes) whether the imposition of a duty would be "fair, just and reasonable" . The adoption of a test cast in those terms does no more than invite attention to what it is that may properly be taken into account in deciding what is fair, just and reasonable and does not suggest any limit to the range of relevant considerations. As Professor Stapleton has said, "[w]ithout more, these are just labels."

In his speech in Caparo Industries Plc v Dickman Lord Bridge of Harwich recognised that reference to what is "fair, just and reasonable" is not useful as a practical test. (It is in Lord Bridge's speech in that case that reference was first made to considering a three stage test of foreseeability, proximity and fairness.) Lord Bridge said of the concepts of proximity and fairness that they: "are not susceptible of any such precise definition as would be necessary to give them utility as practical tests, but amount in effect to little more than convenient labels to attach to the features of different specific situations which, on a detailed examination of all the circumstances, the law recognises pragmatically as giving rise to a duty of care of a given scope".

It is because of the lack of definition of terms like "proximity" and "fairness" that it has been said that the law in this area should develop incrementally . And so it must for as long as no unifying principle emerges. But that is far from saying that the law should develop without explicit recognition of the factors that are considered important in deciding whether there is a duty to take care to avoid pure economic loss. The identification of those factors is essential to any ordered development of the law in this area. In particular, if the matter were to be described in terms of whether to impose a duty would be "fair, just and reasonable" it is essential to identify what are the factors that lead to the application of these epithets. Without that identification the words state no principle, only a qualitative description of the intended result of any and every application of law.

It is not enough to say that compensating those who are injured, deterring wrongdoing or spreading loss are values that are reflected in the law of negligence. They may be. But these do not assist in deciding whether a duty of care exists. They do not assist because each of them is a corollary of a finding that a duty does exist and none, therefore, helps to say whether a duty should be found to exist. Equally, references to the possibility that there are many persons in the same position as a particular plaintiff, or that the losses sustained by a plaintiff and others in like case are very large, do not help any more than do references to floodgates or the like .

As I have said, the search for a control mechanism in addition to foreseeability is driven by at least two considerations - the desire to avoid indeterminate liability and the concern not to establish a rule that will render "ordinary" business conduct tortious. If those are the concerns, then the criterion or criteria devised by the courts should address them directly rather than obscure their significance behind expressions such as "fair, just and reasonable". It may be that there are additional considerations that may have to be taken into account as this area of law develops but for present purposes the two that I have mentioned are critical: whether the liability is indeterminate, and whether the liability is consistent with basic assumptions about the economy in which the conduct takes place. Indeterminate liability?

It is important to understand what is meant by indeterminate liability. It means more than "extensive" or "large". The damage suffered by persons affected by the defendant's negligence may be very large; there may be many who are affected. But neither of those considerations means that the liability is indeterminate. What is meant by indeterminate in the present context is that the persons who may be affected cannot readily be identified. That formulation invites attention to when this identification is to be possible: is it to be possible at or before the time of the negligent act or omission, or is it sufficient if it is possible to identify the class of those affected after the event? I do not think it necessary to say, in this case, whether identification at the later time would suffice. If, as here, it was possible to identify those who would be directly affected by the conduct concerned at the time of the act or omission that is said to be negligent, and it was known to the person alleged to have been negligent that that was possible, then the liability to those persons is not indeterminate.

Here it was possible for the respondent to identify precisely those who would be affected by an outbreak of bacterial wilt caused by its negligence. Those persons could be identified by the respondent because it chose where it would sell the Saturna seed to be grown. It could then readily apply the Western Australian regulation to identify with certainty whether a person would be directly affected by the application of that regulation if the seed that it sold turned out to be affected by bacterial wilt. Those persons would certainly include the growers of potatoes within 20 kilometres of the place where the seed was grown, where the potatoes were intended for sale to Western Australia. Further, I prefer the view that any processor who handled potatoes from that area, and any grower who had potatoes processed with that equipment, should also be included in the class to whom a duty was owed (again, of course, where the potatoes were intended for sale to Western Australia).

It is no answer to say (as the respondent submitted) that many growers may be affected because the respondent sold Saturna seed potatoes to several different growers in different parts of the country. To apply the regulation in the case of each place of sale would increase the number affected but that does not mean that the number is indeterminate.

Other considerations were urged against including processors in the class to whom a duty was owed but, as I say, I prefer the view that they should be included in that class; they are directly affected by the application of the WA regulation. In its terms, the regulation is concerned with entry of potatoes into Western Australia but the potatoes affected are of two kinds - those grown within the affected area, and those processed with certain equipment or at certain premises. It is necessary, in my view, to reflect the inclusion of both in the definition of the class of persons to whom a duty was owed.

The respondent also contended that the class of persons included in those whose crop was processed by equipment used for potatoes from the 20 kilometre area was not capable of ready determination. There is, however, nothing to suggest that potatoes are transported over long distances for processing and it would be possible to say with certainty, before the sale, whether any given individual was or was not a member of the class of growers or processors who would be affected . A known class

Not only is the class of possible plaintiffs capable of definition, and thus determinate, it was a class of which the respondent actually knew. In that respect, this case bears considerable resemblances to Caltex Oil (Australia) Pty Ltd v The Dredge "Willemstad" . Although that decision was criticised, and not followed by the Privy Council in Candlewood Navigation Corporation Ltd v Mitsui OSK Lines Ltd , I consider that Caltex Oil should be followed in this country. That is not to say, of course, that I consider that Caltex Oil identifies the only circumstances in which a person is liable in negligence for pure economic loss. Subsequent decisions in this Court such as Sutherland Shire Council v Heyman , Bryan v Maloney , Hill v Van Erp and Esanda Finance Corporation Ltd v Peat Marwick Hungerfords must all be taken into account. But I do not consider that the decisions in this Court after Caltex Oil suggest that it was wrongly decided.

It will be recalled that the majority of the Court held in Caltex Oil that pure economic loss was recoverable if the defendant had knowledge (or the means of knowledge) that a particular person (not merely as a member of an unascertained class) would be likely to suffer economic loss as a consequence of the defendant's negligence . The class of possible plaintiffs was determinate because it was a class limited to those whom the defendant knew would be likely to suffer economic loss. Here, as the memorandum by the respondent's National Potato Supply Manager demonstrated so eloquently, the respondent knew that persons who grew or processed potatoes for sale to Western Australia would suffer economic loss if the respondent negligently introduced potatoes affected by bacterial wilt to a property within 20 kilometres of those growers or processors. The class of possible plaintiffs was therefore limited in the same way as the class of possible plaintiffs considered in Caltex Oil . An arbitrary class?

It may be said that defining the class of those to whom the duty was owed as those growers and processors to whom the Western Australian regulation would apply directly, is to do no more than seize upon an arbitrary means for limiting liability that happens to be available in this case. But there are at least two answers to that contention. First, the application of any limiting mechanism (whether foreseeability alone, or, in cases of pure economic loss, foreseeability and some other criterion or criteria) will apply tests that will leave some persons within their reach and others beyond it. Any test is, to that extent, an arbitrary one. Secondly, and perhaps more importantly, the application of the Western Australian regulation to define the duty of care is, in this case, the application of a criterion of responsibility of which the respondent knew.

I turn then to the second of the threads that I have said can be identified as important in the development of principles governing liability for negligently caused pure economic loss. An act in trade

The respondent sold seed to the Sparnons because it believed that to do so was in its commercial interest. It hoped to profit from its experiment. To hold it liable to the appellants in negligence would impede persons in the respondent's position from pursuing their economic or commercial interests as they would wish, but that, in itself, is hardly a startling result. There are many respects in which a trader's pursuit of commercial gain must yield if it is to avoid possible liability to a third party.

What, then, does it mean to say that a duty of care to avoid pure economic loss should not be imposed if to do so would cut across accepted bases of economic organisation of this society or would prevent ordinary commercial dealings? Put in either of these ways, the inquiry appears to be value laden. What are the "accepted bases" of the economic organisation of this society? What are "ordinary" commercial dealings? The inquiry can, however, be confined much more narrowly and a negative test applied. In a case such as the present, what would have been the position if the respondent had deliberately (rather than negligently) engaged in the conduct of which complaint is now made? If that deliberate conduct would have been illegal or would have made the respondent tortiously liable to the appellants (or some of them) then it is conduct that would fall outside the boundaries of acceptable commercial dealing. If, by contrast, deliberate conduct would not have been illegal and would not have made the respondent tortiously liable to any of the appellants, there seem very powerful reasons to think that no duty to take care should be imposed in such circumstances. To put the matter another way, if deliberate conduct is neither unlawful nor tortious, why should the same kind of conduct (engaged in carelessly rather than deliberately) be tortious?

Such an inquiry is consistent with the development of the common law in relation to deliberate interference with the trade of another. Much of commercial activity is directed to the pursuit of gain, knowing and intending that success in that pursuit will inevitably harm trade rivals. Yet since Allen v Flood it has been held that malicious interference with another's trade is not actionable if no unlawful means are used. And apart from what might be seen as a temporary diversion in Beaudesert Shire Council v Smith this Court has followed that same path . The principle suggested by Bowen LJ in Mogul Steamship Company v McGregor Gow & Co that intentional conduct "calculated in the ordinary course of events to damage, and which does, in fact, damage another in that other person's property or trade, is actionable if done without just cause or excuse" has not been adopted in this country .

Inquiring whether deliberate conduct would be tortious will often require consideration of the economic torts - procuring or inducing breach of contract and the like. In the present case, however, the conduct of which complaint is made would have been illegal if done deliberately. (It may also have been illegal if done carelessly but I do not stay to examine whether that is so.)

Under the Fruit and Plant Protection Act 1968 (SA) ("the 1968 Act") the importation or introduction into the State of any plant affected by disease might be prohibited by proclamation . In October 1990, bacterial wilt of potato was added to the list of diseases proclaimed under the 1968 Act 1992 and thereafter the importation or introduction into the State of any plant affected by bacterial wilt was absolutely prohibited . Contravention of this prohibition was punishable by fine or imprisonment . (The 1968 Act was repealed in 1992 and the Fruit and Plant Protection Act came into force on 1 February 1993 - after the sale to the Sparnons and after discovery of the outbreak of bacterial wilt on their land. The 1992 Act has broadly similar provisions to those of the 1968 Act. In particular, bacterial wilt has been proclaimed as a disease and the introduction or importation into the State of plants affected by that disease is prohibited.) It follows then, that at the time of the events giving rise to the appellants' claims, the introduction or importation of such seed from Victoria into South Australia was illegal. To hold that the respondent owed a duty of care to some or all of the appellants not to cause pure economic loss by importing affected seed potatoes from Victoria to the Sparnons' property in South Australia would not inhibit the respondent from engaging in conduct that otherwise would have been lawful.

Approaching the matter in this way (by asking what would have been the position if the conduct had been engaged in deliberately) is not inconsistent with the cases in which a duty of care to avoid pure economic loss has been found to exist. Three kinds of cases will serve to make good that proposition. First, if a misrepresentation causing economic loss is made with knowledge of its falsity rather than negligently, the party making the representation is liable for deceit . Secondly, if the dredge in Caltex Oil had deliberately set out to break the pipeline it would have been liable for procuring breach of existing contracts for supply or transport of the hydrocarbons carried in the pipeline. Thirdly, if the solicitor in Hill v Van Erp had intentionally sought to depart from her client's instructions, she would have been guilty of various forms of professional misconduct. The examples could be multiplied.

For these reasons I consider that the respondent did owe a duty to some of the appellants to take care not to cause them pure economic loss. To impose a duty is not to create an indeterminate liability and does not make conduct tortious that is conduct that would not be unlawful or tortious if done deliberately.

The question then becomes to which of the appellants was the duty owed? In my view it was owed to only those of the appellants who would be directly affected by the application of the Western Australian regulation. As I have sought to explain earlier, the respondent could have identified those persons as being those who would be affected by an outbreak of bacterial wilt: the growers and processors of potatoes intended for sale to Western Australia. In the present case that is only Warruga Farms. The land owners did not grow potatoes, and although the Rangara Joint Venture grew potatoes, it sold them to Warruga Farms; the Rangara Joint Venture did not itself sell potatoes for export to Western Australia. Warruga Farms carried on the business of processing the potatoes. Perre's Vineyards owned the processing facility but it stands in no different position from that of the land owners. It was not engaged in the activities to which the regulation applied - growing or processing potatoes intended for sale to Western Australia.

It is important to distinguish between those who were directly affected by the application of the Western Australian regulation and the other appellants. If it is held that a duty is owed to those other appellants who suffered adverse economic consequences because the regulation prohibited Warruga Farms from growing or processing potatoes for export to Western Australia there is, in my view, no sound basis for distinguishing between those other appellants and any other person who suffered adverse economic consequences because of the outbreak of bacterial wilt. The fact that there are family or economic links between the various appellants is not sufficient reason to say that the respondent owed a duty of care to all the appellants. And if it is accepted that the foreseeable occasioning of loss to those other appellants as landowners or landlords is enough reason to impose a duty, what distinguishes their position from that of any other person who indirectly suffered loss as a result of the outbreak? The truck operator whose business was adversely affected because Warruga Farms could not export potatoes and the local store owner who lost trade because Warruga Farms or the Perre family had a bad year are in no different position from the appellants other than Warruga Farms. Each suffers adverse economic consequences because of what has happened. But none (other than Warruga Farms) suffers harm because the Western Australian regulation applies to forbid it selling potatoes to Western Australia. The only persons of whom the respondent can be taken to be aware is the class of growers and processors to whom the Western Australian regulation would apply directly.

As I have already noted, the issue of damages has not yet been tried. It is, therefore, not appropriate to consider any of the questions (perhaps difficult questions) that may arise from the fact that the loss claimed by Warruga Farms is largely a loss of future profit.

The respondent gave a Notice of Contention in which it sought to support the judgment of the Full Court on a number of bases other than those given in the reasons of that Court. I agree with Gummow J, for the reasons he gives, that the matters advanced by the respondent by its Notice of Contention should be rejected.

The appeal by Warruga Farms should be allowed with costs. The appeal of the other appellants should be dismissed with no order as to the costs of those parties in this Court.

CALLINAN J. In this case the appellants claimed damages for economic loss suffered by them as a result of the discovery of an infection of bacterial wilt in a potato crop on a nearby property. In consequence of the application of Western Australian Regulations to properties within 20 kilometres of an infected property the appellants were unable to export any of their potato crop to Western Australia from South Australia. The case comes to this Court on appeal from a decision of the Full Court of the Federal Court which held that the appellants could not recover damages from the respondent who organised the supply of the susceptible seed to the grower of the crop which became infected. The facts

All of the appellants were engaged in various ways in the business of growing or packing potatoes in the South Australian Riverland. The first to eighth appellants (four brothers and their wives) leased from the Crown or owned the land on which potatoes were grown. Warruga Farms Pty Ltd, the ninth appellant (which these natural persons treated as if it were a partnership) was apparently a tenant at will of the land and was the grower of the potatoes. Warruga Farms was also a processor and purchaser of potatoes. The tenth appellant, Perre's Vineyards Pty Ltd (in which the first and second appellants were the shareholders), owned buildings near the farm on which the potatoes were grown. The buildings were occupied and used as washing and packing sheds by Warruga Farms. The 11th to 15th appellants participated in a joint venture known as the "Rangara Joint Venture". Rangara grew potatoes on nearby properties, and sold them to Warruga Farms for washing, packing and resale. The precise nature of the legal relationships between all of these parties is not clear: it seems as if those who used or occupied the properties which were otherwise owned or leased from the Crown, were no more than tenants at will. All of the lands, buildings, sheds and processing plants were situated within 20 kilometres of a property cultivated by the Sparnon Partnership.

Much of the produce of the Rangara Joint Venture and Warruga Farms was customarily exported to Western Australia: in 1992, in the case of the latter, up to 79.2 per cent of its potato crop. Buyers in Western Australia ordinarily paid about $670 per tonne whereas in South Australia the price was much less than that, of the order of $300 per tonne. All of those appellants who were growers grew substantial proportions of their crops with a view to their export to Western Australia. Growers supplying potatoes to persons who in turn would sell them to buyers in Western Australia could be expected to insist upon a price that reflected the higher price that would ultimately be paid in that State. The owners and lessees of the lands who were not growers of potatoes on them had a significant financial interest in the continuing use of the lands for the purpose of growing potatoes on them for export to Western Australia. If potatoes so grown could continue to be exported the capital and rental values of the lands would be enhanced. Not surprisingly therefore, the owners intended and had in mind the purpose that their lands be used for the cultivation of potatoes that could be sold into Western Australia after processing.

The respondent in 1992 controlled 60 per cent of the potato crisping industry in Australia. As part of its operations, it contracted with potato growers for the supply of potatoes suitable for processing into potato crisps. The respondent also conducted research into, and development of new varieties of potatoes for use in its potato crisping operations. The respondent kept in close contact with many of its growers and often chose the varieties to be grown by them and arranged for their acquisition of seed. The respondent imported from Holland some "Saturna" seed which it thought might be suitable for sowing as a winter crop. The attraction of the Saturna potato was that its generally spherical shape was well suited to processing for crisps product.

During the relevant period a department of the Victorian Government conducted a programme for the certification of potatoes and potato seed. In the case of Saturna, certification would only be granted after no fewer than six generations of growth. The respondent entered Saturna in the programme but withdrew it after three generations of growth only. The species did not therefore qualify for certification.

The reason for the decision in the middle of 1990 not to proceed with development of the variety was that the tubers had not measured up to expectations: there were too many variations in size and shape. A decision was taken by staff of the respondent that all of the existing seed should be planted out to exhaust the respondent's stock of the seed. The respondent had not previously used uncertified seed in South Australia for any purpose.

On 26 June 1990 the respondent's national potato supply manager wrote and distributed a memorandum to all of the respondent's state manufacturing managers including the South Australian manager warning against the dangers of bacterial wilt, stressing the need for vigilance in the detection and handling of diseased seed and potatoes, and foreshadowing a real prospect of litigation by growers against the respondent if the latter were seen to be instrumental in the spread of the disease: " The Potato Industry in general is concerned at the increasing rate of incidence of Bacterial Wilt and Potato Cyst Nematode in the past few years. Bacterial Wilt and Potato Cyst Nematode, both of which are 'notifiable diseases' are largely controlled by strict quarantine and long rotations between potato crops ie No potatoes grown for 5 to 10 years. The economic impact on a grower who has the disease on his farm can be disastrous - he is unable to continue to grow potatoes in that part of the farm and unable to sell his infected farm. The trend in the industry is for concentration of activity, in growing, packing and processing, with a consequent increase in risk. We have had contracts for supply with growers who have these diseases , in most states. Recent outbreaks in Qld of Bacterial Wilt, continues the trend. We can anticipate that should it be believed by growers that CCA was instrumental in the spread of the disease to their farm that litigation would be considered. CCA operations must implement adequate controls. Growers still complain of 'old' potatoes being returned in supposedly 'clean' bins. By way of example the Department of Agriculture and Field Officers in the USA where these diseases exist use disposable plastic overshoes etc. The Western Australia Dept of Ag will not allow potatoes sourced within 20 kms of a known outbreak of Bacterial Wilt, into their state. The major cause of spread is through growers buying non certified seed . This often occurs where crops are grown out of season ie Winter harvest and regular certified seed is unavailable at that time of year. Hence there are major outbreaks throughout the Riverina of NSW, Maitland and now Southern Queensland. Whilst seed is the major cause another is Potato Bins. It is in our own interest to maintain vigilance to protect our sources of supply." (Emphasis added)

The respondent selected Mr Tymensen, whose property was located in the Koo Wee Rup Swamp region, to grow Saturna seed. Planting of the crop took place in December 1990. The crop was inspected at least twice by the respondent's supply manager. He observed that the crop appeared healthy.

While Mr Tymensen's crop was in the ground, on 20 and 21 February 1991, a quarterly meeting of some of the respondent's staff was held at Pakenham in Victoria. It was there and then decided that the seed from Mr Tymensen's crop should be used to provide seed for a commercial crop. This was a departure from the respondent's earlier plan to exhaust in one season of planting the remaining Saturna seed.

Mr Tymensen's crop grew to completion and was harvested. The respondent distributed 51 tonnes of the total crop of 86 tonnes to 13 growers in four States, including two tonnes in South Australia. A crop was grown in South Australia and was harvested in December 1991. Six of the respondent's contracted potato growers in South Australia were then approached to grow experimental Saturna crops during the 1992 winter. One of the selected growers to adopt the respondent's proposal to grow Saturna was the Sparnon Partnership on lands within 20 kilometres of the appellants' lands, buildings and plant.

In April 1992, symptoms of "aggressive bacterial wilt" became evident in all six crops being grown in South Australia. Mr Tymensen's property was then inspected by employees of the respondent in an effort to determine the source of the disease. In one corner of the paddock in which the original Saturna crop was grown, "clear evidence" of bacterial wilt infection was found. The respondent, during the hearing before the Full Court, accepted that Mr Tymensen's property was the source of the bacterial wilt that later affected a number of South Australian properties. Crops that had been planted in Queensland and some of the crops in New South Wales and Victoria also became infected by the disease. In all instances, the source of the seed potatoes was Mr Tymensen's property.

The Sparnon properties are as close as three and one half kilometres from the property owned by the Perres. The Warruga Farms property is, at its closest point, three kilometres north. The Rangara property is two kilometres north. But despite this proximity, at no time did the properties on which the appellants grew or handled potatoes actually become infected with bacterial wilt.

The respondent admitted at trial that one of its staff was at some stage telephoned by Frank Perre who raised the possibility that Warruga Farms might become one of the respondent's contract growers. Mr Perre claimed that during that conversation he said that Warruga Farms exported potatoes to Western Australia. The trial judge seems to have accepted this evidence but held that this was information "casually given" and that the purpose of the telephone call was not to convey that information to the respondent. The Full Court upheld those findings .

There was evidence that there were only two commercial growers within 20 kilometres of the Sparnons' property, and, of these, only the appellants exported to Western Australia. However the use to which that evidence may be put is not clear as it may have been admitted against the respondent on the issue of the damages recoverable by the Sparnons only, whose case was, unlike that of the appellants, fully litigated to judgment at first instance.

Infected paddocks cannot be used, as the respondent's memorandum made clear, for the cultivation of potatoes for five years without significant risk. It is very difficult to sell farms on which the disease is present. The treatment of infected paddocks requires the removal of host plants and quarantining. Use of the land is substantially curtailed. Bacterial wilt, and especially the strain detected on the Sparnon property, may remain latent in a plant, and become active again when climatic conditions are favourable to its development. The disease and its potential for damage to the industry were widely known to knowledgeable people in the potato industry, again as the memorandum made plain.

In South Australia, the disease is "notifiable": its occurrence must be reported to the Department of Primary Industries. Yet in that State, the presence of the disease does not mean that produce, including potatoes, from the land on which an outbreak has occurred cannot be sold. But what an outbreak of the disease does, is to attract an embargo upon the export to Western Australia of potatoes grown within a radius of 20 kilometres of a property where it has occurred. This is a consequence of the Regulations made under the Plant Diseases Act 1914 (WA) which prohibit the importation into that State of potatoes grown within such a radial distance of an infected property. Potatoes can only be imported into Western Australia (from States other than Victoria) if an officer from the Department of Agriculture of the exporting State certifies that the potatoes were : "grown on a property situated at least 20 kilometres from a known outbreak of the disease Bacterial Wilt detected within the last 5 years; not, unless otherwise approved by the Director General, harvested, cleaned, washed, graded or packed with equipment or in premises with or in which potatoes, grown within 20 kilometres of a known outbreak of the disease Bacterial Wilt detected within the last 5 years, have been handled; and where contained in bulk bins, placed in bulk bins cleaned of soil and plant material and treated immediately before filling by thoroughly spraying with a 1% solution of formaldehyde."

The application of these provisions made it impossible for the appellants to continue to export potatoes to Western Australia. Earlier proceedings

Following discovery of the disease, and loss of its valuable market in Western Australia, the appellants (as did the Sparnons) commenced proceedings against the respondent for negligence. The Sparnons also alleged causes of action based on s 14 of the Sale of Goods Act 1895 (SA) and s 71 of the Trade Practices Act 1974 (Cth). The damages were potentially very large. The appellants claimed $8.933m at trial. The owners of the Sparnon property also claimed in respect of their losses that the State of South Australia and the State Department of Primary Industries had negligently failed to investigate the history of the original Saturna seed and diagnose that it was infected with bacterial wilt. The claims against the State and the Department were dismissed at trial and were no longer pursued.

The trial judge von Doussa J, by consent, embarked upon a separate trial confined to the issue of liability of these appellants' claims against the respondent. The Sparnons' claim was litigated on all issues at trial.

Although the primary judge found that the respondent was in breach of its duty of care owed to the Sparnons, the appellants were unsuccessful. His Honour found that the respondent owed no duty of care to Warruga Farms on the basis that: "the factors relied [upon] ... do not create a special relationship of proximity with [the respondent] which takes [the] claims in negligence out of the general rule that damages are not recoverable for purely economic loss."

The trial judge found that the respondent had never heard of Perre's Vineyards and Rangara. His Honour referred to Caltex Oil (Australia) Pty Ltd v The Dredge 'Willemstad' and quoted a passage from the reasons of Gibbs J in that case in which his Honour said that foreseeability of loss may not be enough to make that loss recoverable but that in some exceptional cases, and taken with other matters, it may give rise to a claim for economic loss.

The trial judge gave judgment in favour of the Sparnons in negligence for a sum of money.

The appellants appealed to the Full Court of the Federal Court (constituted by O'Loughlin, Branson and Mansfield JJ) against the dismissal of their claims in negligence against the respondent . There was no appeal against the judgment in favour of the State of South Australia. The respondent appealed against the judgment in favour of the Sparnons. The Sparnons did not appear when the appeal came on for hearing. The Full Court nonetheless dismissed the respondent's appeal against the judgment in favour of the Sparnons.

The Full Court was of the view that the appellants' claims were within a new category of cases not covered by the decision and principles stated in Caltex or any of the subsequent cases in this Court in which pure economic loss has been recovered.

After reviewing the authorities the Full Court said : "It is, in our view, appropriate to elicit from the above decisions, the additional observation that despite the absence of any anterior dealing between [the respondent] and [the appellants], it would be relevant to the existence or otherwise of a relationship of proximity to found the necessary duty if [the respondent's] conduct in relation to the Sparnons carried with it some circumstance involving direct reliance by, or an assumption of responsibility towards, or control of events directed to producing an outcome for, the [appellants]. Whether such considerations, if they exist, amount to a refinement of the 'identity of interests' or 'common adventure' descriptions that have been used in some cases, or are simply another way of expressing the same, but more general, concept as discussed by Gaudron J and Gummow J in Hill v Van Erp , is not critical to resolution of the present appeal. The subsequent decision of the High Court in Esanda Finance Corporation Ltd v Peat Marwick Hungerfords , which dismissed a claim for damages for negligent misstatement against a firm of auditors is also consistent with our above observation. ... [I]t is, in our view, difficult to elevate the relationship between the [appellants] on the one hand and [the respondent] on the other to the category of 'special' as that expression has been applied in the High Court."

Their Honours were influenced by the absence of anterior relationships between the respondent and the appellants. They also commented on the absence of evidence of how many growers there were within a radius of 20 kilometres of all growers who planted Saturna seed, and, of these, where their potatoes were washed or packed, and how many of them were exporters to Western Australia. They thought that the losses sustained here were not distinguishable in kind from losses that might be sustained in adopting measures to avoid contamination. All of the appellants fell outside, their Honours concluded, any ascertained class; and, even if they were to be regarded as being members of an ascertainable class, there was no circumstance such as, for example, a customary sharing of washing facilities by the Sparnons with the appellants to require the respondent to take precautions for the protection of the appellants. Their Honours concluded that the appellants' detriment was quite different in quality from that suffered by the plaintiff in Caltex.

The appellants' appeal to the Full Court of the Federal Court therefore failed. The appeal to the High Court

The appellants' grounds of appeal to this Court which were pressed were that the intermediate court erred in that it: 1. "failed to find the instant case came within the principles enunciated in Caltex Oil (Aust) Pty Ltd v The Dredge 'Willemstad' for the purposes of deciding whether a duty of care not to cause economic loss was owed by the respondent to the appellants"; 2. "failed to equate causing an apprehension of injury with causing actual injury for this purpose"; 3. "failed to correctly consider both the actual knowledge and constructive knowledge possessed by the respondent's servant Hughes of the economic harm which Hughes' negligent acts would have on the appellants as individuals. In particular it failed to take into account Hughes' actual and constructive knowledge of the appellants, their location within 20 kilometres of the Sparnon property and the potato exports conducted by the appellants to Western Australia"; 4. "failed to adequately consider the evidence of Hughes' knowledge by dismissing it as evidence of knowledge 'casually given' to Hughes and therefore not knowledge acquired by the respondent"; 5. "failed to find that the appellants were members of a determinate class of potential claimants"; 6. "failed to appreciate the similarity in nature of the economic loss caused by the respondent's negligence to that which would have been caused by actual physical contamination, namely the appellants' inability to grow or process export potatoes for Western Australia on the appellants' property for a period of five years from the date of the detection of the outbreak of bacterial wilt".

The respondent has filed a notice of contention but consideration of it may be deferred.

The appellants essentially presented two arguments in this Court: that the nature of the damage suffered by the appellants was not "pure economic loss" but that it was properly characterisable as a physical effect on property falling just short of physical injury, but indistinguishable in principle from physical injury, or, it flowed from a threat or apprehension of actual physical damage and was therefore sufficiently analogous to loss consequent upon actual physical damage to justify recovery; alternatively, if the damage suffered was properly to be characterised as "pure economic loss", a duty of care was owed because of the existence of special circumstances.

In Caltex this Court held that, although as a general rule damages are not recoverable for foreseeable economic loss which is not consequential on injury to person or property, damages may be, and were there recoverable in a case in which the defendant had knowledge, or the means of knowledge that a particular plaintiff would be likely to suffer economic loss as a consequence of the defendant's negligence. Although the principle emerging from Caltex may be stated in those broad terms, its application in any particular case requires a careful marshalling of all the relevant facts and circumstances, and has on occasions been frankly acknowledged as involving the weighing of relevant policy considerations.

In Caltex Gibbs J reviewed the earlier cases, and discussed some of the policy considerations and other reasons which had led the courts in the past to hold against liability. His Honour accepted the possibility of illogicality in the distinction between recovery for economic loss and recovery for material loss, but expressed the same view as Lord Denning that the distinction was founded in common sense : "It may be right to say ... that the distinction between recovery for economic loss and recovery for material loss is illogical, but that does not mean that the decisions that have drawn that distinction were erroneous, because the law aims at practical justice rather than logical consistency. However, I am in respectful agreement with Lord Denning MR in SCM (United Kingdom) Ltd v W J Whittall and Son Ltd that the distinction is not lacking in common sense. If a person committing an act of negligence were liable for all economic loss foreseeably resulting therefrom, an act of careless inadvertence might expose the person guilty of it to claims unlimited in number and crippling in amount. For example, if, through the momentary inattention of an officer, a ship collided with a bridge, and as a result a large suburban area, which included shops and factories, was deprived of its main means of access to a city, great loss might be suffered by tens of thousands of persons, but to require the wrongdoer to compensate all those who had suffered pecuniary loss would impose upon him a burden out of all proportion to his wrong. Similarly, the driver of a vehicle which collided with a pylon carrying electric power lines in an industrial area might dislocate the work of dozens of factories. It is true that under modern conditions some claims arising from physical injury or material damage can be very large in amount - for example if a passenger train were derailed. Nevertheless, the extent of claims for loss that is purely economic is likely to be very much wider than that of claims arising out of physical injury and material damage. Further, a law which imposed a general duty to take care to avoid causing foreseeable pecuniary loss to others would ... interfere greatly with the ordinary affairs of life. There are sound reasons of policy why economic loss should not be treated in exactly the same way as material loss." His Honour's conclusions were stated in this passage : "In my opinion it is still right to say that as a general rule damages are not recoverable for economic loss which is not consequential upon injury to the plaintiff's person or property. The fact that the loss was foreseeable is not enough to make it recoverable. However, there are exceptional cases in which the defendant has knowledge or means of knowledge that the plaintiff individually, and not merely as a member of an unascertained class, will be likely to suffer economic loss as a consequence of his negligence, and owes the plaintiff a duty to take care not to cause him such damage by his negligent act. It is not necessary, and would not be wise, to attempt to formulate a principle that would cover all cases in which such a duty is owed; to borrow the words of Lord Diplock in Mutual Life & Citizens' Assurance Co Ltd v Evatt : 'Those will fall to be ascertained step by step as the facts of particular cases which come before the courts make it necessary to determine them.' All the facts of the particular case will have to be considered. It will be material, but not in my opinion sufficient, that some property of the plaintiff was in physical proximity to the damaged property, or that the plaintiff, and the person whose property was injured, were engaged in a common adventure."

Stephen J accepted that policy considerations dictated the need to look beyond mere foreseeability of loss or damage as a test for liability for economic loss. His Honour said : "The need is for some control mechanism based upon notions of proximity between tortious act and resultant detriment to take the place of the nexus provided by the suggested exclusory rule which I have rejected. Its precise nature and the extent to which it should restrict recovery for purely economic loss must depend upon policy considerations just as does the conclusion that for cases of economic loss such an additional control mechanism is necessary. Both in actions for negligent misstatement and in products liability actions based upon negligence, the particular fact situations encountered are likely themselves to provide material out of which formulations limiting the extent of liability may be fashioned; Hedley Byrne and Rivtow Marine respectively provide examples of this process in these two areas. But in the general realm of negligent conduct it may be that no more specific proposition can be formulated than a need for insistence upon sufficient proximity between tortious act and compensable detriment. The articulation, through the cases, of circumstances which denote sufficient proximity will provide a body of precedent productive of the necessary certainty; the gradual accumulation of decided cases and the impact of evolving policy considerations will reflect 'the courts' assessment of the demands of society for protection from the carelessness of others' - per Lord Pearce in Hedley Byrne reiterated by Lord Diplock in Dorset Yacht Co v Home Office . It was Lord Pearce in Hedley Byrne who explained the divergence between the law of negligence in word and that of negligence in act in terms of the quite special characteristics of words as the instrument of negligence . Economic loss possesses many of the characteristics which Lord Pearce attributed to negligence by word and the need which his Lordship recognized for proximity as a precondition of liability for negligence by word applies equally to all cases of recovery for purely economic loss. Some guidance in the determination of the requisite degree of proximity will be derived from the broad principle which underlies liability in negligence. As Lord Atkin put it in a much cited passage from his speech in Donoghue v Stevenson the liability for negligence 'is no doubt based upon a general public sentiment of moral wrongdoing for which the offender must pay' . Such a sentiment will only be present when there exists a degree of proximity between the tortious act and the injury such that the community will recognize the tortfeasor as being in justice obliged to make good his moral wrongdoing by compensating the victims of his negligence. Again, as Lord Morris said in the Dorset Yacht Case courts may have recourse to a consideration of what is 'fair and reasonable' in determining whether in particular circumstances a duty of care arises ; so too, I would suggest, in determining the requisite degree of proximity before there may be recovery for purely economic loss."

Mason J emphasised the undesirability of the multiplication of claims by a large number of members of a general class of persons for a virtually indeterminate liability : "It is preferable, then, as Mr P P Craig suggests in his illuminating article, 'Negligent Misstatements, Negligent Acts and Economic Loss' that the delimitation of the duty of care in relation to economic damage through negligent conduct be expressed in terms which are related more closely to the principal factor inhibiting the acceptance of a more generalized duty of care in relation to economic loss, that is, the apprehension of an indeterminate liability. A defendant will then be liable for economic damage due to his negligent conduct when he can reasonably foresee that a specific individual, as distinct from a general class of persons, will suffer financial loss as a consequence of his conduct. This approach eliminates or diminishes the prospect that there will come into existence liability to an indeterminate class of persons; it ensures that liability is confined to those individuals whose financial loss falls within the area of foreseeability; and it accords with the decision in Rivtow ."

Jacobs J adopted a somewhat broader approach which does not seem to have attracted the support of other members of the Court then or subsequently. In his Honour's opinion, where foreseeable economic loss arises from a physical effect on the plaintiff's property there is no bar to recovery on the ground only that the loss is economic. In a passage upon which the appellants rely for their first argument Jacobs J said : "A question of central importance in the present case is whether the physical effect in this context is limited to actual physical injury. In my opinion it is not. Person or property may be injured not only physically but also by physical effect thereon short of physical injury: eg by an act or omission which prevents physical movement of a person or which prevents physical movement or operation of property. The damages for immobilization of property may frequently be quantified as the cost of mobilizing the property and the loss of the use of the property during its immobilization. Such damage may be called pecuniary or economic loss. However, it is an error to concentrate attention on the question whether a particular loss is pecuniary or economic. Rather it is necessary to examine the circumstances of the loss. If the loss arises from the physical effect of an act or omission on the person or property of a plaintiff and that physical effect is one which was foreseeable and that foreseeability gives rise to a duty in the defendant to take reasonable care to avoid that physical effect, it is no answer to the plaintiff's claim for damages that his loss was pecuniary or economic. Nor is there any need or place for such a concept as has been expressed in the term 'parasitic damages' ."

Murphy J expressed a quite different view which left little or no room for the limitation of damages or liability of ordinary people or corporations on grounds of policy except in the case of public authorities or corporations who his Honour thought might require some special protection or immunity .

In Bryan v Maloney Mason CJ, Deane and Gaudron JJ adopted the same sort of approach as commended itself to Stephen J in Caltex . Their Honours said : "The cases in this Court establish that a duty of care arises under the common law of negligence of this country only where there exists a relationship of proximity between the parties with respect to both the relevant class of act or omission and the relevant kind of damage. In more settled areas of the law of negligence concerned with ordinary physical injury to the person or property of a plaintiff caused by some act of the defendant, reasonable foreseeability of such injury will commonly suffice to establish that the facts fall into a category which has already been recognized as involving a relationship of proximity between the parties with respect to such an act and such damage and as 'attracting a duty of care, the scope of which is settled' . In contrast, the field of liability for mere economic loss is a comparatively new and developing area of the law of negligence. In that area, the question whether the requisite relationship of proximity exists in a particular category of case is more likely to be unresolved by previous binding authority with the consequence that the 'notion of proximity ... is of vital importance' . As Stephen J indicated in Caltex Oil (Australia) Pty Ltd v The Dredge 'Willemstad' , it is the 'articulation', in the different categories of case, 'of circumstances which denote sufficient proximity' with respect to mere economic loss, including 'policy considerations', which will gradually provide 'a body of precedent productive of the necessary certainty'. Inevitably, the policy considerations which are legitimately taken into account in determining whether sufficient proximity exists in a novel category will be influenced by the courts' assessment of community standards and demands . One policy consideration which may militate against recognition of a relationship of proximity in a category of case involving mere economic loss is the law's concern to avoid the imposition of liability 'in an indeterminate amount for an indeterminate time to an indeterminate class' . Another consideration is the perception that, in a competitive world where one person's economic gain is commonly another's loss, a duty to take reasonable care to avoid causing mere economic loss to another, as distinct from physical injury to another's person or property, may be inconsistent with community standards in relation to what is ordinarily legitimate in the pursuit of personal advantage . The combined effect of those two distinct policy considerations is that the categories of case in which the requisite relationship of proximity with respect to mere economic loss is to be found are properly to be seen as special. Commonly, but not necessarily, they will involve an identified element of known reliance (or dependence) or the assumption of responsibility or a combination of the two ."

In Hill v Van Erp , Dawson J and Toohey J held that a disappointed beneficiary who was deprived of a testamentary disposition by reason of a solicitor's negligence in inviting the beneficiary's husband to attest the will could recover against the solicitor because of the duty of care arising out of the proximity between the beneficiary and the solicitor .

An important consideration in that case was that there was no prospect of indeterminate liability: the class of persons to whom the duty was owed was readily ascertainable. Giving effect to the duty did not diminish any competitive advantage a solicitor might have or supplant or supplement other legal remedies. Furthermore, the acknowledgment of the existence of the duty did not disturb any general body of rules constituting a coherent body of law. Gaudron J and Gummow J were both of the view that an important factual matter arguing in favour of liability was the misuse of effective control exercised by the solicitor, and that such a factor would always be relevant and important in determining whether a case was of such a kind as to justify an award of damages for pure economic loss.

In Esanda Finance Corporation Ltd v Peat Marwick Hungerfords , this Court held that a financier was not entitled to succeed on the facts pleaded against auditors of a borrower corporation for losses from transactions which it claimed it had entered into in reliance upon audited accounts and the accompanying auditor's report. The financier alleged that mandatory accounting standards applied and that the auditors had fallen short of those standards in conducting the audit. The financier claimed that it was a member of a class of persons whom the auditors foresaw or ought reasonably to have foreseen might reasonably have relied upon the accounts and the report.

The Court (Brennan CJ, Dawson, Toohey, Gaudron, McHugh and Gummow JJ) unanimously rejected the financier's claim but for reasons that were not identical. Brennan CJ thought it critical that, among other things, the circumstances justifying recovery should be such that the defendant knew, or ought reasonably to have known that the information would be communicated to the plaintiff, either individually, or as a member of an identified class, and that it would be so communicated for a purpose that would be very likely to lead the plaintiff to enter into a transaction of a kind that he did enter; and further, that it would be very likely that the plaintiff would enter into such a transaction in reliance upon the information and thereby risk economic loss if the information were to turn out to be untrue or unsound .

Dawson J thought it important in a case of pure economic loss that, in addition to foreseeability of harm, there be a special relationship between the parties which might be described as a relationship of proximity . Both Dawson J and McHugh J thought it relevant that there was a statutory regime in force to regulate the conduct of auditors and the contents of any report that they made or auditing work that they did .

Toohey and Gaudron JJ were of the view that whether a duty of care to take reasonable steps to avoid economic loss to another exists, depends upon a relationship of proximity which may be found in special categories of cases only, such as those in which there is an expectation, rather than mere reliance, on the part of the person who has suffered the loss . So too, an assumption of responsibility for providing information in a context in which it is known or ought reasonably to be known that it may be acted upon for a serious purpose will be, in a case such as Esanda , of importance. Their Honours thought that other relevant questions were whether the defendant possessed any special expertise or knowledge, or had special access to information not available to the recipient of the information, and the reasonableness or otherwise of the conduct of the recipient in acting upon the information without further inquiry .

McHugh J had regard (among other things) to wider considerations of policy such as the impact of prolonged and multiple actions against auditors upon the courts, the insurability or otherwise of the risks, the magnitude of the losses of which they would be productive if realised, the availability of other civil proceedings, the possibility of criminal proceedings against the auditors, and the public interest in ensuring that there be auditors ready, willing and able to perform their professional duties at a reasonable cost . Gummow J was of the opinion that the allegations, if established, would not be sufficient to found a duty of care, and that there was not a sufficiently close relationship between the financier and the auditors to warrant any imposition of liability upon them .

In Candlewood Navigation Corporation Ltd v Mitsui OSK Lines Ltd the Judicial Committee was of the opinion that there was not a sufficient commonality of statement of principle in the respective judgments of Gibbs J, Stephen J and Mason J in Caltex to enable a clear ratio to be extracted from that case and that therefore they did not think they could derive any assistance from it . Since Candlewood however there have been a number of cases in Australia, both in this Court and in other courts , in which Caltex has been considered and applied or distinguished.

The cases subsequent to Caltex in this country show that all judges are united in their opinions that, for policy reasons, there is a need for a control mechanism to limit the availability of relief for pure economic loss so that commerce, providers of services, courts and society generally will not have to bear the burden and uncertainty of incalculable claims by a mass of people whose identity or very existence may be unknown to the defendant. It is not surprising, having regard to the different factual situations in which pure economic loss has been suffered and will no doubt be suffered in the future, and the frank judicial acknowledgments that have been made of the relevance of public policy and social issues, that the principles governing or controlling the mechanisms to limit liability have not always been stated identically.

The different path which this Court has followed in Caltex and the ensuing cases to which I have referred makes it unnecessary to consider Caparo Industries Plc v Dickman in the House of Lords and Hercules Managements Ltd v Ernst & Young in the Supreme Court of Canada (applying Caparo ) which may in any event perhaps be taken to depend upon the special statutory framework in which auditors' reports are prepared, matters which both Dawson J and McHugh J thought of relevance in the Australian auditor's case of Esanda .

In her essay "Duty of Care Factors: a Selection from the Judicial Menus" , Professor Stapleton discusses some of the difficulties which have confronted the courts in trying to find, and express one clear principle which judges may readily apply in cases of pure economic loss. There is much I think to commend her view that : "while the listing of these judicial menus of sound factors relevant to the duty issue help unmask the substantive determinations being made by judges in this field, they cannot operate as some sort of mechanical guide as to how a novel case will be decided in the future. ... [A]t the end of the day, even if judges agree on the relevant factors to be weighed in the individual case, different judges may well place different weight on competing factors and do so quite reasonably." It should be made clear however that the determination of a claim for pure economic loss is not a merely discretionary matter: it requires the application of the principles stated in Caltex and the subsequent cases in this Court to the various factual situations as they arise in the courts.

And it must be accepted that this is an area of the law in which the courts should move incrementally and very cautiously indeed. It is not yet possible to identify a bright line of demarcation between those cases of pure economic loss in which damages are recoverable and those in which they are not. The law is still developing in the somewhat piecemeal fashion that Stephen J predicted in Caltex : "As the body of precedent accumulates some general area of demarcation between what is and is not a sufficient degree of proximity in any particular class of case of pure economic loss will no doubt emerge; but its emergence neither can be, nor should it be, other than as a reflection of the piecemeal conclusions arrived at in precedent cases."

I turn now to a consideration of the factors which in combination I think relevant in this case and which establish a sufficient degree of proximity, foreseeability, a special relationship, determinacy of a relatively small class, a large measure of control on the part of the respondent, and special circumstances justifying the compensation of the appellants for their losses.

The respondent played a leading role in the industry. It was involved in all phases of it, from the introduction of particular varieties of seed, and the encouragement of growers to sow that seed, to the acquisition of the produce of that seed, its processing into a manufactured commodity and the distribution of that commodity in the community. In short, its role and position in the market were commanding ones. This role of itself tended very much to place the respondent in a special relationship to growers and handlers of potatoes.

The respondent was in effective control of the particular operation which led ultimately to the imposition of the embargo upon the appellants' properties . That operation was the selection of the Saturna seed and its invitation to Mr Tymensen and then the Sparnons to sow it.

I am of the view that the appellants were members of a determinate class. As Hayne J points out, the respondent effectively chose where the seed was to be grown. By force of the Western Australian Regulations, if bacterial wilt did occur, then all of those operating in the industry and involved in the provision of land for, and the cultivation and preparation of potatoes for the Western Australian market, within 20 kilometres of the occurrence of the disease were capable of being directly affected by the respondent's actions. The evidence shows that there were relatively few growers in fact of potatoes within 20 kilometres of the Sparnons' property. Furthermore, only 26 growers in the whole country were asked to plant this seed. The respondent, having regard to its dominating position in the industry, must have known, or at least it ought to have known, of the special attractions of the Western Australian market and the likelihood that growers and handlers of potatoes such as the appellants would wish to take advantage of it. Furthermore, the respondent was well aware, as the memorandum of 26 June 1990 shows, that the trend in the industry was for concentration of activity, in growing, packing and processing, with a consequent increase in risk. Accordingly the respondent actually foresaw as being within a class of people likely to be adversely affected, packers and processors of potatoes, including those who were also growers, or handlers and land owners or facility owners closely connected with them such as those appellants as were not growers.

The geographical propinquity of the property to which the respondent caused the Saturna seed to be introduced (the Sparnons' property) to the appellants' property is relevant .

So too, the commercial propinquity, that is to say the facts that the appellants and the respondent were both involved in the same industry in the same year and had been so involved for some time, is relevant. Both this and the preceding matter to which I have referred bespeak, in a real sense, proximity.

I think it relevant that the respondent in this case had an especially heightened awareness of the dangers of bacterial wilt and the strict attitude of the Western Australian authorities towards that disease. Those portions of the memorandum of 26 June 1990 which I have emphasised make this proposition graphically clear. The effects of an embargo go beyond slight harm. Indeed, as the memorandum also makes clear, for a particular grower or processor the consequences of an outbreak within 20 kilometres could be devastating as they were here, effectively putting the appellants out of business.

In this case the respondent assumed a risk of which it was well aware or should have been aware . It made two important, conscious decisions against the background of its awareness of the potential consequences of an outbreak of bacterial wilt. It chose to withdraw the seed from the certification programme and to invite growers to plant it when it knew that the best way of avoiding the risk was to ensure that certified seed be used. There would have been no inconvenience to the respondent in taking appropriate steps to avoid risk of injury to the appellants. All that the respondent needed to have done was to have encouraged and arranged the use of certified or other proved seed, or to have persisted with the certification programme for the Saturna seed.

The harm that was done to the appellants was done in relation to their participation in a national commodity market. It is notorious that commodity markets are fragile (as the evidence here indicates) and in particular are very vulnerable in modern times to contamination both prospective and actual. Interstate trade in commodities, as well as in other products, is to be encouraged, and it is apparent that there was in place in this case a co-operative scheme for the benefit of the industry involving authorities in Victoria, South Australia and Western Australia and perhaps even elsewhere.

What the respondent was doing here was undertaking an experiment. The respondent's officers referred to the use of the seed on the Sparnons' property in those terms and clearly so regarded it. An experiment almost always involves risks and those risks needed evaluation in light of the dangers which the respondent's own officers were stressing in writing .

The appellants were rendered powerless to abate, or to prevent the occurrence of the loss to which they were subjected . In no way did they act illegally, improperly, or unreasonably or without regard for their own interests.

It was not suggested in argument before this Court that there was any statutory penalty or other sanction which might be imposed upon the respondent or that there was any statutory or regulatory regime to govern the way in which the respondent relevantly acted in effectively controlling the planting of the seed.

Although it could not be said that the appellants had any particular expectations with respect to the selection of the seed by the respondent, the appellants were entitled to assume that those who were involved in and played a leading role in their industry and might distribute or arrange for the use of seed which if infected could damage the appellants' business and properties, would be responsible in the way in which they caused or permitted a particular seed to be sown.

What happened to the appellants here was not the result of merely legitimate, competitive, commercial activity .

I do not regard the likely number of growers at risk actually or potentially as a result of the respondent's activities to be so numerous as to be likely to give rise to a crippling burden upon the respondent. In a country such as Australia it is open I think, for a court to take judicial notice of the likelihood that generally speaking the cultivation of potatoes is not a market garden activity but will usually be a broadacre activity not involving a very large number of growers, especially growers exporting to Western Australia, and in an area within a radius of 20 kilometres from an infected property. The evidence here is that the area of Mr Tymensen's property was 69 hectares. The Sparnons' holding was 409 hectares of which at least 14 hectares were sown with potato seed in 1989. The Perres' property was 807 hectares of which more than a hundred hectares were used for growing potatoes.

The imposition of liability upon the respondent would not impose an impediment in the way of ordinary commercial activity in the potato industry .

What the respondent did went considerably beyond careless inadvertence and resulted from conscious decisions carrying with them obvious risks.

What happened to the properties (the lands, the plant and equipment, and the leaseholds or tenancies) of the appellants here may not have been actual physical damage but it came very close to that. "Blight" is a familiar concept in compensation and town planning cases . In a sense, the amenity of a property is very much akin to a physical attribute of a property. A particular activity on parcel "A" may adversely affect the amenity of parcel "B" although the two parcels do not adjoin each other. This may occur for example, as a result of a massive and overwhelming development on parcel A rendering obsolescent a building on parcel B; or a particular activity on parcel A, although not constituting an actionable nuisance, may make the outlook from parcel B much less attractive, or subject it to noise pollution. A similar comparison might be made in respect of plant and equipment the use of which has been effectively sterilized by force of a Regulation, in the same way as the infliction of physical damage to it might disable it. Such effects are very similar to actual physical damage and are not logically readily distinguishable from physical damage. Absent negligence or infringement of legislation the causing of blight will not ordinarily be actionable. Such effects are very similar in impact to the negligently caused effects upon the appellants' properties of the outbreak of bacterial wilt on the Sparnons' property leading to the Western Australian embargo, and consequentially, in a real sense, involve the imposition of a blight upon their properties by way of a significant reduction in the utility and productivity of them, and accordingly their value. I regard this too therefore as a relevant consideration.

To grant the appellants relief in this case would not undermine any principle of law or liability which might otherwise be invoked, or which might provide a basis upon which the respondent might otherwise resist the appellants' claims .

To hold this respondent liable represents no departure from previous case law.

The considerations which I have outlined bring this case within the principles stated in Caltex and the subsequent cases in which it has been applied and outweigh the relatively few countervailing considerations to which the respondent points in resisting the claims.

One of the major touchstones in a case of this kind will always be reasonableness , or as it has sometimes been put, proportionality . I do not regard the damages which are likely to be available to the appellants as being unfair, or unreasonable, or disproportionate in all of the circumstances of this case.

The Court became aware after the hearing that there was legislation which prohibited the importation or introduction into South Australia of any plant affected by bacterial wilt . There had been no reliance on that legislation in the Federal Court. At the very least the existence of that legislation as well as the stringent Western Australian requirements may serve to emphasise the importance that the market and governments attach to uncontaminated produce. But because the parties did not argue the case by reference to this legislation, I do not take it into account in reaching my decision.

The respondent correctly submits that this is not a case of a common adventure: it is not a case in which the appellants can be shown to have relied upon any statement, or act, or abstention from doing an act by the respondent . The respondent also urges that its responsible officers did not know of the existence of the appellants except for Frank Perre and did not in fact know that the appellants, except perhaps for Warruga Farms, exported potatoes to Western Australia. There was no direct injury to the appellants' property and no infection by bacterial wilt of crops grown on it. The appellants were not deprived of an opportunity of selling their potatoes other than to buyers in Western Australia.

To each of these matters, undoubtedly relevant as they may be, there is a clear answer. Liability is not confined to a case of common adventure. The appellants were entitled to expect that a person such as the respondent would act carefully and responsibly in carrying out an experimental activity that had a real and acknowledged potential to cause grave harm to the appellants. The appellants fell within a class of which the respondent was or should have been aware: growers and processors of potatoes for possible export to Western Australia operating within 20 kilometres of the Sparnon property. In a case of economic loss there will frequently be no "direct" or "physical injury" to property. The appellants' chances of sale of potatoes other than into Western Australia provide no consolation for the appellants' loss of their valuable Western Australian market. Accordingly neither singly nor collectively do the respondent's arguments provide an answer to the matters to which I have referred in detail earlier. Those factors make it clear in my view that what happened here was foreseeable, and that there was sufficient proximity between the appellants and the respondent to give rise to liability on the part of the respondent. The case is an exceptional one of a special relationship between the parties arising out of the unusual combination of circumstances operating here. The appellants' "class" was a relatively small and determinate one. The respondent owed a duty of care to the appellants and failed to discharge a duty of reasonable care by taking the risks and acting in the way in which it did. The Full Court of the Federal Court failed to take into account not only several of the factors which are important in this case, such as the experimental nature of the respondent's activity, the commanding position of the respondent in the industry, the risks associated with the use of a new, (in this country) uncertified seed, the fragility of the market, the relative ease of avoiding the risk, and the great harm done to the appellants, but also the cumulative effect of these, and the other matters I have listed. For those reasons I would allow the appeal.

It is unnecessary to express any opinion on the appellants' first argument that the fact that there may have been a species of harm to, or impact upon the appellants' properties indistinguishable from, or very similar to physical harm entitled the appellants on that account alone to recover.

The respondent has filed a notice of contention. Much of what is sought by it, effectively findings of unforeseeability, is already covered by what I have so far said and should be rejected. The notice also takes up a matter adverted to by the trial judge, that is, as to the characterisation of what had occurred at a meeting of employees of the respondent at Pakenham in Victoria as negligence. However, what occurred at that meeting was not the negligence relied on by the appellants to establish their claim. It is true that the pleadings do not in terms define the appellants' case as well as they might have but there is no doubt that the issue of causative negligence upon which it turns was fully litigated: the causing or permitting of the experimental cultivation of an uncertified, relatively recently imported variety susceptible to bacterial wilt within 20 kilometres of the appellants' land and business operations. That is of course only a broad particular of the negligence and has to be considered in the context of the whole case.

What else is fatal to the notice of contention is that it really seeks to re-litigate in this appeal the finding of negligence in favour of the Sparnons who are not parties to this appeal. That finding was made in proceedings which were heard together with these. No finding other than that the respondent was negligent in arranging for the Sparnons to grow Saturna was open in either proceedings. The evidence leading to that result was largely uncontradicted.

The appeal should be allowed with costs and the matter remitted to the Federal Court for a hearing on the issue of damages.

I would make some observations about damages. The appellants are not in identical positions. There is obviously a difference between the situation of the owners of land, its cultivators when they are not the owners, and the processors of the crop. Although all were sufficiently proximate to the respondent, and sufficiently directly affected by its negligence to enable them to make claims in the particular circumstances of this case, the processors because of their close connexion with the affected lands and producers, their exploitation of the Western Australian market, and the quarantining of the processing plant by operation of the Western Australian Regulations; and, the land owners who were not actual cultivators, because of their intention and purpose in causing their lands to be used to grow potatoes for export to Western Australia and the economic impact upon their lands of the statutory embargo; the proof and assessment of damages in respect of each of them may call for the application of different approaches and care to ensure that there is no double counting. It may even be possible that one or more appellants will be unable to prove any substantial damages.

Care does need to be taken in deciding whether to conduct separate trials of different issues. It sometimes happens that they may turn out to be productive of the disadvantages of delay, extra expense, appeals and uncertainty of outcome which they are intended to avoid. In tort cases in which damage is the gist of the action, it will generally be undesirable to accede to requests for them, or to order them, unless all parties accept that compensable damage has been sustained by the plaintiffs or applicants as the case may be.

The order I would make is appeal allowed with costs and further order that the matter be remitted to the Federal Court for trial on the issue of damages.