It has frequently been said that the modem law of damages dates back no farther than the middle of the last century. To be sure, long before that time the legal profession had come to recognize the function of damages, namely, to give "compensation and satisfaction for some injury sustained." 2 Blackstone, Commentaries *438. And yet, the middle of the last century has rightly been hailed as the turning point in the evolution of the law of damages, for only then were decisive steps taken by judges and text writers to bring rationality and predictability into the substantive law of compensation and to develop efficient procedural techniques for translating theory into practice.2
To enable the reader to appreciate the full meaning of this new phase in the development of damage law, a short survey of the preceding period may be helpful. From the beginning of trial by jury, the amount of damages 'was a "fact" to be found by the "petit" jury selected from the neighborhood where the transaction in litigation had occurred. The trial judge, "usually a stranger sent out from London, with no knowledge of the affair in controversy except from the pleadings, could seldom feel justified in correcting or overturning the findings of the neighbor witnesses." C. T. McCormick, Law of Damages 25 (1935). See also Tooley and Preston's Case, 3 Leon. 150, 74 Eng. Rep. 599 (1587).3
But "outrageous and excessive" verdicts occurred so frequently that the development of techniques for controlling this grave judicial risk became a matter of necessity. Expectations created by promises were often vitiated completely by an inadequate award of damages or magnified unreasonably by an excessive award. Contracting parties began to attempt to bring this element of uncertainty under private control by incorporating into their agreements clauses providing for the payment of a specific sum of money in case of default. These clauses were honored, at first even where they provided for onerously high payments; only gradually, under the influence of equity, did there emerge the modern distinction between valid provisions for liquidated damages and unenforceable penalties. McCormick, supra, at 599 et seq. See further Section 4. But private control was not enough. A system of judicial control emerged slowly through a laborious process of trial and error. The first step taken was not very successful. As far back as 1300, a proceeding appeared that enabled the party aggrieved by the petit jury's verdict, particularly if it was excessive, to petition a jury of 24 knights for a retrial. If the retrial was granted, a writ of attaint issued. But this system had a rather serious shortcoming: if the aggrieved party was successful in the trial, not only was the "false" verdict replaced by a new verdict, but the members of the first jury were punished severely for having violated their oath to find a correct award based on their knowledge of the facts. Small wonder that the attaint came to be replaced by other methods of controlling the juridical risk, such as the granting of a new trial before a second petit jury, a step that was forced upon the common law courts by the intervention of equity.4
The widening of the power of courts to set aside "false" verdicts was accompanied and put on a firmer basis by the gradual evolution of rules designed to standardize the quantum of recovery. These rules, at first, were used only in advising the jury; eventually they took the form of binding instructions, and slowly the doctrine emerged that a misdirection by the court was grounds for a new trial. Attempts to control the juridical risk by standardizing the quantum of damages made their first appearance in cases dealing with transactions as commercially significant as loans and sales, and it is striking that these early efforts at rationalization sought, almost without exception, to introduce order and predictability into the law of damages by limiting the amount a disappointed promisee could recover. Thus, for example, in a line of cases dating back to Lord Mans- field's decision in Robinson v. Bland, 2 Burr. 1077 at 1086, 97 Eng. Rep. 717 at 722 (1760), the damages for nonpayment of a debt were limited to interest, no matter how great the real loss suffered by the creditor. Similarly, under the famous rule of Flureau v. Thornhill, 2 W. Blackstone 1078, 96 Eng. Rep. 635 (1766), the liability of the vendor of real estate who, without being guilty of bad faith, failed to make title to his purchaser was restricted to the return of the down payment and sometimes other expenses, i.e., to the reliance interest. Loss of profits (the expectation interest) remained unprotected. In the law of sales, it is true, early cases like Gainsford v. Carroll, 2 B. & C. 624, 107 Eng. Rep. 516 K.B. (1828), infra p. 1129, recognized the promisee's right to his expectancy and thus helped to safeguard the profit motive in an area of economic activity where this seemed especially important. But even here, the promisee's damages were limited by the standardized rule restricting his recovery, in the event of nondelivery or nonacceptance, to the difference between the contract price and the market price prevailing at the time and place of delivery appointed in the contract.5
Apart from these rather specialized rules dealing with the types of contracts just enumerated, we find in the case law before 1850 little more than occasional statements that damages must be the "natural" or "necessary" result of the promisor's breach. The development of an overall principle "by which the judges could justify keeping a firm hand upon amounts awarded for breach of contract, so as to confine such awards within the risks which the judges would believe to be in accord with the expectation of business men" (McCormick, supra, at 563) was slow in coming indeed. Not until Hadley v. Baxendale, 9 Ex. 341, 156 Eng. Rep. 145 (1854), supra p. 106, did a court succeed in devising persuasive and highly generalized formulae for determining the quantum of recovery for breach of contract, formulae that have served as models for the instruction of juries in countless cases ever since. In the language of the court:6
Where two parties have made a contract which one of them has broken, the damages which the other party ought to receive in respect of such breach of contract should be such as may fairly and reasonably be considered either arising naturally, i.e., according to the usual course of things, from such breach of contract itself, or such as may reasonably be supposed to have been in the contemplation of both parties, at the time they made the contract, as the probable result of the breach of it.
9 Ex. at 354, 156 Eng. Rep. at 151.8
In keeping, perhaps, with the spirit of the earlier damages rules described above, the great generalizations of Hadley v. Baxendale were also understood, by contemporary lawyers, to be aimed essentially at limiting, rather than expanding, the promisee's recovery. This restrictive view of the great case, which prevailed for some time both in the case law and the treatises, is strikingly illustrated by the following passage from J.D. Mayne, Treatise on the Law of Damages 8 (2d ed. 1872):9
. . . In the case of contract the measure of damages is much more strictly confined than in cases of tort. As a general rule, the primary and immediate result of the breach of contract can alone be looked to. Hence, in the case of non-payment of money, no matter what the amount of inconvenience sustained by the plaintiff, the measure of damages is the interest of the money only. So where the contract is to deliver goods, replace stock, or convey an estate, the profit which the plaintiff might have made by the resale of the matter in question cannot in general be taken into account; nor the loss which he has suffered from the fact of his ulterior arrangements, made in expectation of the fulfilment of the bargain, being frustrated. The principle of all these cases seems to be, that, in matters of contract, the damages to which a party is liable for its breach ought to be in proportion to the benefit he is to receive from its performance. Now this benefit, the consideration for his promise, is always measured by the primary and intrinsic worth of the thing to be given for it, not by the ultimate profit which the party receiving it hopes to make when he has got it. A bottle of laudanum may save a man his life, or a seat in a railway carriage may enable him to make his fortune; but neither is paid for on this footing. The price is based on the market value of the thing sold. It operates as a liquidated estimate of the worth of the contract to both parties. It is obviously unfair, then, that either party should be paid for carrying out his bargain on one estimate of its value, and forced to pay for failing in it on quite a different estimate. This would be making him an insurer of the other party's profits, without any premium for undertaking the risk.
In the preceding section on Specific Performance the suggestion was made that a restrictive or negative approach to the availability of the remedy has, in certain areas at least, gradually been replaced by an expansive and, perhaps, more liberal approach. In much the same way, the restrictive interpretation of the true meaning of the rule of Hadley v. Baxendale, current in the nineteenth century, has in this century gradually been replaced by a more expansive one. See The Heron II, infra p. 1157. This enlargement in our conception of compensable harm has done its part to help dissolve the idea of a narrow but strict liability that lay at the heart of the classical (that is to say, late nineteenth-century) view of contractual obligation. The broader the scope of his responsibility for remote or "consequential" damages, the more deeply the promisor will be drawn into the affairs of his contractual partner and the more likely he is to become, in effect, a guarantor of their success. To this extent, the liberalization of the rule in Hadley v. Baxendale belongs to the same general movement of ideas as our growing acceptance of the so-called duty.to negotiate in good faith — a duty that the law imposes even before there is a contract, let along a breach by one of the parties. These two rules, operating, so to speak, at opposite ends of the life or career of a contractual relationship, both tend to transform the parties from self- interested entrepreneurs with only limited responsibilities for one an- other's welfare into joint-venturers engaged in a common undertaking premised upon reciprocal duties of support. How far this transformation has been (or should be) carried, and the way it affects our conception of the shifting boundary between tort and contract, are questions to which we shall return.11
On this last issue — the relation of contract to tort — a further word is perhaps in order. In 1847, when he published his treatise on damages, Sedgwick already assumed, without discussion, that contract damages are meant to provide "simply. . . compensation," being in that respect unlike tort damages, which may include a punitive element. That twin proposition has been repeated so many thousands of times that one begins to wonder whether anything more than a sort of ritual incantation is involved.12
The exact location of the dividing line between contract and tort is of course a mystery that the high priests of the legal profession have always been concerned to preserve and protect from public view. The cat was let out of the bag, however, in a moment of unusual candor, by Lord James of Hereford in Addis v. Gramophone Co., Ltd.,  A.C. 488 (H.L.). The manager of the defendant's Calcutta office had been wrongfully dismissed. Because of the "harsh and humiliating" manner of the dis- missal, the jury had included in its verdict a substantial sum in addition to the salary and commissions to the end of the term to which plaintiff was concededly entitled. This was reversed in the Court of Appeal, which was in turn upheld in the House of Lords (although Lord Collins felt that the jury verdict should have been allowed to stand). Lord James of Hereford, concurring with the majority on the proposition that punitive damages are never recoverable in a contract action, said:13
. . . My Lords, I must say if I had arrived at a different conclusion I should have been subject to some feeling of remorse, because during many years when I was a junior at the Bar, when I was drawing pleadings, I often strove to convert a breach of contract into a tort in order to recover a higher scale of damages, it having been then as it is now, I believe, the general impression of the profession that such damages cannot be recovered an action of contract as distinguished from tort, and therefore it was useless to attempt to recover them in such a case. That view, which I was taught early to understand was the law in olden days, remains true to this day.
 The A.C. at 492.15
The proposition that exemplary or punitive damages are never recoverable in a contract action has suffered some erosion since the Addis case. The erosion, it may be, has gone further in this country than in England. At all events, a considerable number of American cases can be found in which the courts, without going through the ritual of converting a breach of contract into a tort" (which is, of course, another way of doing it), have allowed such damages in "contract" actions. Most of the cases involve situations in which the defendant's behavior could be, and was, characterized as "wanton," "willful," "reckless" and so on. Collections of such cases can be found in 11 Williston §§1340, 1341.16
The other, and equally questionable, side of the "simply . . . compensation" coin is the implicit claim that a monetary award, calculated according to standard damage formulae, will in fact adequately compensate an injured promisee for the loss he has incurred. We have seen that the rules for measuring contract damages were, to begin with, undercompensatory (and may very well have been designed to be so). Perhaps a good deal of the difficulty 'that the courts have had with damage theory over the past hundred and thirty years is attributable to the fact that less- than-compensatory formulae have, somehow, had to be squared with the "compensation" idea, without, at least overtly, abandoning either the formulae or the idea.17
 For the further development of the rule, see Washington, Damages in Contract at Common Law (Pt. 2), 48 L.Q. Rev. 91 (1932); C.T, McCormick, Law of Damages 205 et seq. (1935).18
 In their attempt to reduce the juridical risk of uncertain jury awards by devising an overall formula for measuring contract damages, the courts were greatly aided by the work of a number of able text writers, both American and English, who, in tum, were strongly influenced by the development of the French law. Some of the great cases contain express references to Pothier (translated in 1806) and the French Civil Code (with which the legal profession had become familiar through Sedgwick's classic treatise on damages first published in 1847).19
 Note, The Expanding Availability of Punitive Damages in Contract Actions, 8 Ind. L. Rev. 668 (1975); Note, Punitive Damages in Contract Actions — Are the Exceptions Swallowing the Rule?, 20 Washburn L.J. 86 (1980).