Concept
The Section 351 Test
Causation and foreseeability
The Section 351 Test
Contract law gives an injured party the benefit of its bargain: it seeks to place that party where performance as promised would have left it.1 That measure rests on the principle that makes contract law valuable at all: private persons can enforce the beneficial deals they make with one another.2 Section 351(1) of the Restatement (Second) of Contracts trims the measure in a single sentence: damages are not recoverable for loss that the party in breach did not have reason to foresee as a probable result of the breach when the contract was made.3
The rule descends from Hadley v. Baxendale, which allowed damages that may fairly and reasonably be considered either to arise naturally, “according to the usual course of things,” from the breach, or 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.”4 Section 351(2) restates those two routes. A loss is foreseeable if it follows from the breach in the ordinary course of events,5 or as a result of special circumstances, beyond the ordinary course of events, that the party in breach had reason to know.6 Later sections take up each route. This section covers the four filters that apply to both.
The single sentence of § 351(1) imposes four conditions, and each phrase does separate work. The first is the standard of knowledge: the breacher must have had reason to foresee the loss, which asks what it should have known and does not ask what it actually thought. The second is timing: the court looks to the moment the contract was made and ignores what the parties learned afterward. The third is degree: the loss must have been a probable result of breach, which means more than a possible one. The fourth concerns the object of foresight: the breacher must have foreseen a loss of a certain kind, and the cases divide on how closely it must have foreseen the loss's size. A student who reads the sentence as one undivided question, “was it foreseeable?”, will miss the points on which cases turn.
All four conditions protect the promisor's chance to bargain over risk. A promisor that sees a risk at formation can reject the deal, raise the price, include a clause limiting or excluding recovery, buy insurance, or take extra precautions; without reason to know that losses of a type are likely, it has no meaningful chance to do any of these.7 Each phrase of § 351(1) guards that chance in its own way, and the subsections below take the phrases in order.
Reason to Foresee
The operative word is reason, and it sets an objective standard. Foreseeability concerns what the breacher should have foreseen if it had reasonably contemplated the course of likely future events, and not what it actually foresaw.8 The rule therefore requires reason to know and not necessarily actual knowledge: even a particularly dense defendant answers for the damages it should have foreseen.9 Actual knowledge still suffices, since a party that knew certainly had reason to know.
The alternative shows why the law chose this standard. If liability turned on what the defendant actually foresaw, the best evidence would be the defendant's own account, and every defendant sued after a breach would say that the loss never crossed its mind. The rule would reward the promisor who took care to learn nothing. An objective test replaces that unprovable inquiry with one the record can answer: what did the breacher know when the contract was made, and what would a reasonable party in its position have inferred from it?
Hadley v. Baxendale shows the standard working in a defendant's favor. The carrier's delay caused the mill's lost profits in fact, yet the court held the carrier not accountable, because it had not been told that this was the mill's only shaft and had no way of knowing that a delay would leave the mill idle.10 The question was what the carrier's information would have told a reasonable carrier, and the answer was that it said nothing of an idle mill.
The standard asks for reason to foresee and for nothing more. A stricter version of the test demanded that the circumstances of the contract show that the breacher tacitly assumed the risk of liability, in addition to the probable consequences being reasonably foreseeable.11 A later section addresses that version; for now, hold the question at reason to foresee.
The foresight at issue concerns the consequences of a breach and leaves the chance that a breach will occur out of account. A defendant cannot avoid damages by showing that, at formation, she had no reason to know she would decide to breach. The court asks, hypothetically, what losses she should have considered had she considered the effects of a breach like the one that occurred.12 If sincerity at signing defeated liability, almost every promisor would escape, and a rule meant to limit damages would excuse breach altogether.
Time of Contracting
The clock stops at formation. Under § 351(1), the breacher must have had reason to foresee the loss as a probable result of breach when the contract was made,14 so what it learned later, during performance or on the eve of breach, does not matter. The time of contracting, and not the time of breach, must necessarily be the point at which foreseeability is gauged.15 Hadley used the same date when it asked what both parties could reasonably be supposed to have contemplated “at the time they made the contract.”16
Formation controls because that is when the parties assumed their respective obligations and determined the nature of the performance they reasonably contemplated.17 Events during performance, such as changes in market conditions, do not affect the initial question; a court asks whether the breaching party should have contemplated the type of risk when the bargained-for exchange occurred.18 The principle is that a breaching party should not answer for a loss it could not reasonably have anticipated as a consequence of breach when it entered the contract.19
The timing rule also protects the bargaining that § 351 exists to preserve. A promisor can reject the deal, raise the price, limit its liability, insure, or take extra precautions only if it knows the risk, and it can do so only before it is bound.20 If notice given after signing could enlarge liability, a plaintiff could create exposure unilaterally by announcing special needs just after the contract was made and then collecting for a risk that no price reflected. The rule also assigns a task to the party who alone knows the unusual facts. That party must guard against the loss before contracting, either by carrying an inventory of replacement parts or by telling the other side about its special circumstances when the contract is made. The mill owner in Hadley did neither, and recovery was barred.21
Marcus & Co. v. K.L.G. Baking Co. applies the timing rule to a buyer's claim for lost subcontract profits. A seller of goods answers for losses that would probably result in the ordinary course of things from a breach under the special circumstances known to the parties when the contract was made, because the law ordinarily treats such losses as reasonably within the contemplation of both parties.22 The converse follows. The court quoted an earlier authority: if the seller was not told of a subcontract at the time of the sale, then its knowledge that the buyer meant to resell generally, or notice of a subcontract “given to him subsequent to the date of the contract, will not render him liable for the buyer's loss of profits on such subcontract.” In such cases the special injury was not within the seller's contemplation as the probable consequence of breach, so the seller owed no compensation for it.23
The date itself can be disputed when a contract's terms are fixed in stages. Fort Pitt, a bridge fabricator, argued that under Hadley whatever knowledge it received after the contract was made could not expand its liability. Torrington answered that Fort Pitt should be charged with what it knew in November, when the June delivery date was agreed, and that an experienced bridge fabricator must then have realized that any delay beyond August would jeopardize the pouring of the concrete and force postponement of the work until spring.24 Both sides accept that the date of contracting controls; they disagree about which date that is. The dispute shows that the rule is only as clear as the date on which the relevant term was made.
Do not confuse the foreseeability date with the date used to fix market price. Most case law and commentary sets market price as of the date substitute performance could be effected, and the competing possibilities create costly uncertainty when a contract is repudiated in advance.25 That is a question about measuring damages, and it leaves untouched the question what the breacher had reason to foresee when it contracted.
Probable, Not Merely Possible
The third condition is degree. Section 351(1) denies damages for loss the breacher had no reason to foresee as a probable result of breach,27 and Hadley used the same word when it spoke of damages reasonably supposed to have been in the parties' contemplation “as the probable result of the breach of it.”28 The word matters because a loss that has already happened can always be called possible in hindsight. If possibility were enough, every loss traced to the breach would pass the limit, and the limit would remove nothing from a claim. Requiring a probable result gives the limit work to do: it separates losses a sensible party in the breacher's position would have expected from losses that arrived through an unlikely run of events.
The standard has two edges, and a court must respect both. At the upper edge, the loss must be more than conceivable. That edge keeps a promisor from having to bargain over every remote consequence of breach: when a loss was probable, the defendant should bargain over it or pay for it, and when it was remote, the plaintiff should raise it in negotiation or bear it. At the lower edge, the loss need not be nearly inevitable. Some courts find foreseeability only when a loss was nearly inevitable, which reads the “ordinary course of events” as the inevitable and treats everything else as unusual. Once the plaintiff has mentioned the loss, however, a defendant will have difficulty persuading a court that it was unforeseeable merely because it was not inevitable.29
Newsome v. Western Union Telegraph Co. shows the upper edge. The plaintiff sought damages from a telegraph company over a telegram ordering whisky, and the company's agent had been told what the whisky was for. The court held that damages that are uncertain and speculative, or that are not the natural and probable result of the breach, are too remote to be recovered, that damages may not rest on a mere conjectural probability of future loss or gain, and that “something more than a possible result must appear.”30
The case also rests on a second ground that a careful answer keeps separate. The defendant was a public agency, compelled to accept the telegram at a price fixed by the North Carolina Corporation Commission, so it could not be said to have contracted with the claimed damages in mind merely because its agent knew the whisky's purpose.31 Newsome therefore combines a speculative loss with a defendant that had no chance to bargain, and it is a mistake to cite it as a pure foreseeability holding.
A 1905 decision, Coppola v. Kraushaar, draws the same line from the other direction. A seller failed to deliver two gowns ordered for a bride, the wedding was postponed, the bride broke the engagement, and the disappointed suitor sued to recover the five hundred dollars he had spent on the wedding.32 The court wrote that the parties must have had in contemplation that the wedding would never occur if the seller failed to furnish the dresses on time, and that it was “hardly to be contemplated, in the absence of specific warning, that she would forever refuse to wed” if they were late. The damages were “too remote.”33 A canceled wedding was imaginable, but nothing the seller knew made it probable, and the phrase “specific warning” points to what would have changed the result: disclosure at contracting.
Step two does most of the work, because it separates a loss that follows directly from the breach from one that needs a chain of favorable events. Other regimes phrase the degree of likelihood in their own words, and the sales and international sections below return to the comparison. On a common-law contract problem, apply the Restatement's word, probable.
Kind of Loss, Not Amount
The fourth condition concerns what the breacher must foresee, and that is a loss, with no mention of a sum.35 The working rule is that the breacher need not have foreseen the amount or exact nature of the loss or the precise way it would occur; it is enough that the breacher reasonably have foreseen losses of the kind and scale that would likely result from the breach.36 A seller of fabric to a clothing factory, for example, need not know the exact transactions that depend on the cloth. It is sufficient that the seller knew or had reason to know the buyer would use the cloth to make clothing, which the buyer would sell in the ordinary course to its customers.37
The rule takes this shape because foresight of a precise sum is impossible at formation. Nobody knows at signing how long a delay will last, what the market will do meanwhile, or how the plaintiff's later dealings will turn out. A promisor told the kind of exposure a contract carries can still raise the price, insure, or limit liability, and none of those steps requires knowing the loss to the dollar. A rule demanding foresight of the exact amount would excuse nearly every defendant from nearly every consequential claim.
The same latitude covers the way a loss comes about: the defendant need not predict the precise route the loss takes, only that a loss of that type was to be expected.38 Decisions such as Williams v. Gray Group, L.L.C. apply that latitude.
Basic Capital Management, Inc. v. Dynex Commercial, Inc. shows how much a lender must know. The court agreed that a general knowledge of a prospective borrower's business does not give a lender reason to foresee the probable results of refusing to lend. But it found no authority for Dynex's position that the consequences of breaching a loan commitment are unforeseeable unless the lender knew, when the commitment was made, not only the nature of the borrower's intended use of the money but also the specific venture in which the borrower meant to engage.39
An older case points the same way. In National Bank of Cleburne, a bank breached its agreement to lend a mill $14,000 to buy wheat, and the mill sued for the profits it would have made on resale. An appellate commission reversed the judgment for the mill because there was no evidence that the bank knew the mill intended to resell the wheat. Had there been such evidence, the bank would not have needed to know to whom the mill intended to sell it.40 The two statements together carry the lesson. Knowing the nature of the borrower's intended use makes the kind of loss, lost resale profit, foreseeable; knowing only that the borrower is in business does not.
The kind-of-loss reading does not make the size of the loss irrelevant, and the cases divide. Reason to anticipate the type of loss generally suffices to make damage foreseeable. Some cases, though less commonly, require reason to know the amount of the loss, and under that reading a plaintiff recovers lost profits on a resale only if the defendant had reason to know how large the profit would be.41 Treat the split as a majority tendency with contrary authority and not as a settled line. The “scale” language in the working rule is where the two readings meet.
On the amount-sensitive reading the excess drops out for the same bargaining reason that supports the kind-of-loss rule.42 A repairer told of a resale can price the job against a typical resale profit, but it cannot price against a windfall created by the buyer's unusual need, and the owner can disclose that need when the contract is made. Where a loss is foreseeable but its size dwarfs what the breacher earned under the contract, a separate limit may apply, which the section on disproportionate compensation takes up below.
Sales of goods add remedies with their own vocabulary. The Code tries to balance remedies evenly, so sellers generally have access to the same categories of damages as buyers, each complemented by incidental damages for expenses reasonably incurred as a result of the breach.43 A buyer's right under UCC § 2-502 to recover identified goods it has paid for is a different remedy, one that vests when the goods are identified to the contract.44 UCC § 2-715 receives its own section below.