Skip to main content

Contract Clause


In a flashing aperçu Sir Henry Maine observed that "the movement of progressive societies has hitherto been a movement from Status to Contract." In feudal systems a person acquired a fixed, social status by birth, one's legal rights and duties being determined thereby for life. The decline of feudalism was a fading away of the status system in favor of personal rights and duties based largely on contractual relationships. Obligations imposed by ancestry gave way to obligations voluntarily undertaken. Generally thereafter a person's place in society depended upon success or failure in covenants with respect, for example, to wages, raw materials, farm and industrial goods, or artistic talent. In such a setting it is crucial that agreements be dependable—not merely to promote the individual's security and mobility but for the good of a society that relies for its sustenance upon a vast network of voluntary, contractual relationships. Thus Article I, section 10, of the Constitution, reflecting in part unfortunate experience under the articles of confederation, forbids inter alia state laws "impairing the obligation of contracts." In the federalist #44, james madison, observed that such laws "are contrary to the first principles of the social compact and to every principle of sound legislation." In his view "the sober people of America" were "weary of fluctuating" legislative policy, and wanted reform that would "inspire a general prudence and industry, and give a regular course to the business of society."

Indeed the sanctity of contracts was deemed so vital to personal security that in fifty-five years following the Supreme Court's first contract clause decision (fletcher v. peck, 1810), twenty-two states put such provisions in their own constitutions. With one exception each of them was included in the state's bill of rights. Prior to 1810 four states had already done this. All of them protected contracts generally (per Fletcher), not merely private contracts as in the northwest ordiance. Plainly in john marshall's day and long thereafter his Court's broad view of the contract clause was widely accepted—along with freedom of religion, and fair trials—as one of those restrictions on government "which serve to protect the most valuable rights of the citizen."

Obviously those who thus equated property rights and civil liberty were—like the Founding Fathers and the marshall court—disciples of john locke. He had taught that property and liberty go hand in hand; that neither thrives without the other; that to protect them both as indispensable to life itself is the reason for government. Generations later, in a radically changed economic setting, some Americans came to believe that property hampers liberty. Inevitably then (having forgotten Locke) they would misunderstand both the founders and our early judges—Lockians all. Thus the Progressive movement convinced itself and its heirs that the Marshall Court had erred in holding the contract clause applicable to state, that is, public, covenants and that in so holding the judges had revealed a pro-property bias. Both of these views—derived largely from Fletcher and Dartmouth College v. Woodward (1819)—seem erroneous. The first rests on the strange idea that unambiguous language of the Constitution means not what it plainly says, but rather something else, because of the supposed intent of its authors. (Of course authors' intent may be a proper key to the meaning of ambiguous terminology, but that is a very different matter.)

Had the constitutional convention of 1787 wanted the clause to cover only private agreements, that is, those between individuals, it need only have said so. The Continental Congress had done just that in the Northwest Ordinance: "… no law ought ever to be made, or to have force in the said territory, that shall in any manner whatever, interfere with or affect private contracts.…" Six weeks later, rufus king moved to include its private contract approach in the Constitution. Following a brief discussion of possible ramifications of such a provision, it was dropped. A few days later, at the suggestion of the Committee of Style, the Constitutional Convention adopted the contract clause, which refers comprehensively to "contracts" without qualification. Nothing in our record of the proceedings explains the change of mind or the change of terminology. But this is certain: not a word there or in The Federalist even hints that the founders were concerned only with private covenants—that they thought a state should be free to violate its own agreements. alexander hamilton would later observe: "It is … impossible to reconcile the idea of a [state] promise which obliges, with a power to make a law which can vary the effect of it." Hamilton, of course, had been a member of the Constitutional Convention.

Long before John Marshall became a judge, Justice james wilson, in chisholm v. georgia (1793), had asked rhetorically: "What good purpose could this constitutional provision secure if a state might pass a law impairing the obligation of its own contracts, and be amendable, for such a violation of right, to no controlling judiciary power?" This from one who had been perhaps the second most important leader of the Constitutional Convention. Justice william paterson, too, had been influential at the Convention. Years before Fletcher in a similar case, van horne ' s lessee v. dorrance (1795), he had held that a state could not impair its own contractual obligations. So did the highest court of Massachusetts in Derby v. Blake (1799) and in Wales v. Stetson (1806). Fletcher was not without significant judicial precedent.

The argument that the contract clause does not mean what it says rests essentially on the proposition that the crucial contract problem in late eighteenth-century America was erosion of private contract obligations by debtors' relief legislation. No doubt this was a vexing and well-known difficulty. Yet surely it is no ipso facto basis for excluding related problems plainly covered by explicit constitutional language. State negligence with respect to state obligations was after all a matter of experience. Even if it were known that the Framers intended the written words to embrace only private contracts, judges could not properly adopt that view. For those who ratified can hardly be said to have ratified something other than the words of the document. To hold otherwise is to undermine the basic premise of a written constitution. As the Marshall Court put it in orthodox manner in Dartmouth: "This case being within the words of the Contract Clause, must be within its operation likewise, unless there be something in the literal construction so obviously absurd or mischievous, or repugnant to the general spirit of the instrument, as to justify those who expound the constitution in making it an exception.…" No such basis for an exception having been discovered, the Supreme Court ever since has found the contract clause applicable as written to contracts generally, whether public or private.

A related problem in Fletcher concerned the scope of the term "contract." The Georgia legislature had sold and granted to speculators millions of acres of public land. A subsequent legislature had repealed the grant on the ground that it had been obtained by bribery. Meanwhile part of the land had been conveyed to innocent third-party purchasers. The issue in Fletcher was whether the initial grant entailed obligations protected against impairment by the contract clause. The Court responded affirmatively. Of course, in modern usage a grant is not a contract, but that does not solve the problem. For it is quite clear that in the late eighteenth century the term "contract" had far broader connotations than it does today. As Dean roscoe pound has explained:

Contract was then used, and was used as late as Parsons on Contracts in 1853 to mean [what] might be called "legal transaction." … Not merely contract as we now understand it, but trust, will, conveyance, and grant of a franchise are included.… The writers on natural law considered that there was a natural legal duty not to derogate from one's grant.… This is the explanation of Fletcher, … and no doubt is what the [contract clause] meant to those who wrote it into the Constitution. ["The Charles River Bridge Case," Massachusetts Law Quarterly 27:19–20.]

In sum, in the context of the times a grant included an executory, contractual obligation of the grantor not to violate the terms of his grant.

On this and the public contract aspect of Marshall's opinions one finds no criticism or disagreement in a random selection of twelve legal treatises published before 1870, including thomas m. cooley's famous Constitutional Limitations (1868). Twelve years later, however, Cooley's General Principles of Constitutional Law (1880) took a somewhat critical stand. The change apparently reflected attacks upon the Marshall Court by C. M. Hill, R. Hutchinson, and J. M. Shirley writing separately in the American Law Review and the Southern Law Review from 1874 to 1879. In due course the Progressives would pick up these charges that the Marshall Court had erred, and they would add that "the great Chief Justice" was in fact "a stalwart … reactionary," a servant of property interests. It seems no coincidence that the Hill-Hutchinson-Shirley attacks germinated in an era (1865–1873) when nearly sixty percent of laws challenged under the contract clause were held invalid—an all-time high. Many such cases of course "favored" business interests and thus tended to offend Progressives. The result was that John Marshall became for them a villain.

Fletcher, of course, upheld the property claims of innocent, third-party purchasers. The alternative would have been to sustain the property claims of the innocent people of Georgia. Either way the judges would be deciding in favor of some, and against other, property interests. Either way innocent people would suffer. One fails to see how Fletcher can be said to reveal a property bias. Was there, however, bias of another sort in deciding for the ultimate buyers rather than for the initial owners? Far from exceptional, the choice was informed by a long settled (and still prevailing) rule of Anglo-American equity jurisprudence. Although a fraudulent purchaser takes a good title at law, it is subject to cancellation by a chancery decree. Thus in a clash between a cheating buyer and his innocent victim the latter prevails. But Fletcher involved a clash between the innocent victim and an equally innocent, subsequent purchaser for value. With the equities thus in balance and the social interest in security of transactions on the side of the purchaser in possession, the chancellor does not intervene (the victim's recourse being an action for damages against the fraudulent party). In short, Marshall and his Court read the contract clause in the light of a long familiar rule of equity.

In the Dartmouth case, a group of philanthropists had received a public charter to create a college in New Hampshire. Later the state tried to take over and govern the school contrary to the charter provisions. Marshall's Court, following Fletcher, held that the charter was a contract which the state was not free to violate. Viewed narrowly the case was won by the college trustees, but they had no beneficial interest in the college property. They won on behalf of the donor-philanthropists (presumably deceased) and generations of future students.

The Progressive response is that the Dartmouth decision was a crafty gambit purposefully designed for an ulterior purpose: protection of corporation charters from legislative interference. That it was highly successful is demonstrated, we are told, by the enormous growth of corporate enterprise thereafter. This is make-believe. Justice joseph story in Dartmouth pointed out that no state need grant irrevocable or nonamendable charters—that the power to amend or revoke may be reserved. Damage resulting from failure to do so can hardly be held a fault of the judges. In fact reservation of power to alter corporate charters became widespread after Dartmouth and was not unknown before. (See reserved police power.)

Fletcher and Dartmouth are not pro-property, but rather pro-transaction, cases. They mean that when judges find no clearly overriding public interest such as they found in gibbons v. ogden (1824) they will not disturb the contractual arrangements, that is, the transactions, by which women and men conduct their affairs—be they philanthropists (Dartmouth) or land speculators and farmers (Fletcher). As Marshall put it, "the intercourse between man and man would be very seriously obstructed, if this principle be overturned." Incidentally, by killing New York's restrictive steamboat law Gibbons too promoted transactional freedom.

An inclination toward unfettered private activity was deep in the temper of the times. Americans were on the make. They had escaped the old-world fetters: king, feudal aristocracy, and established church. They were the "new men." A vast geographical frontier invited initiative and ingenuity. The standard of living was low, but natural resources were plentiful. These conditions put a high premium on private, developmental effort. Such was the setting in which contract clauses found their way into bills of rights along with other basic protections then deemed indispensable to personal freedom and social well-being.

If the Marshall Court found that the Constitution forbade reneging on state obligations, it also recognized that public agreements raised special problems justifying a special rule of strict construction. In providence bank v. billings (1830) Rhode Island had chartered the Providence Bank "in the usual form" with no reference of stipulation concerning taxation. Later, when the state enacted a bank tax, Providence Bank argued that a power (taxation) which might be used to destroy its charter was foreclosed by implication. The Court demurred: "as the whole community is interested in retaining [the power to tax] undiminished, that community has a right to insist that its abandonment ought not to be presumed, in a case in which the deliberate purpose to abandon it does not appear." Obviously the mere grant of a corporate charter for an ordinary banking operation could not rationally be held to imply an immunity from routine, nondiscriminatory taxation.

Marshall's Jacksonian successors under Chief Justice roger b. taney followed the established path with two modifications: a temporary enlargement of the rule of strict construction, and a decision that a state may not by covenant fetter its power of eminent domain. The former occurred in charles river bridge co. v. warren bridge co. (1837). Massachusetts had authorized private investors to build, operate, and maintain a public drawbridge in exchange for toll rights for a period of forty (later extended to seventy) years. Before that period expired the state authorized a competing, in effect toll-free, bridge only yards away from the original facility. Was the state free thus to jeopardize the revenue of the first bridge, or did the forty-year provision implicitly preclude such interference? It must have been clear at the outset to all concerned that investors would not provide, maintain, and operate for forty years a public facility, if the state were free at any time to disrupt their only source of compensation. Surely in these circumstances the Marshall rule of strict construction was satisfied; the state's "deliberate purpose" to permit unimpeded toll collection for the period in question seems obvious.

The Taney Court did not repudiate—indeed it purported to follow—the Providence Bank rule of strict construction. In fact, it simply ignored the "deliberate purpose" aspect of that rule, and substituted an incompatible principle derived from English precedents: "nothing passes by implication in public grants." (Thus did Harvard College lose part of its endowment.) Justices Story and smith thompson dissented on implied agreement grounds. Justice john mclean agreed with them on the merits, but thought the Court lacked jurisdiction. In substance Charles River was a 4–3 decision—although five of the Justices had been appointed by President andrew jackson, the other two by his Jeffersonian predecessors.

The majority position—exalting form over substance—would have permitted construction of an adjacent, toll-free bridge immediately after construction of the first one. Yet surely no court would so decide. If this be true, the Taney rule against implied agreements must be untenable. The Court seems rarely to have used it, having returned long ago to the Marshall approach. See, for example, northwestern fertilizing co. v. hyde park (1878): "Nothing is to be taken as conceded but what is given in unmistakable terms, or by an implication equally clear" (emphasis added). The major upshot of the Taney Court's stricter rule of construction was that those who covenanted with a state took care to secure elaborately explicit commitments.

Charles River Bridge exudes liberalism, stressing as it does the social interest in progress. Property rights, the Court proclaimed, must not impede developing technology. Old turnpike charters, for example, should be construed "strictly" lest they block new railroads. But no such clash of old and new was at issue before the Court. The real problem was that after some forty years the Massachusetts legislature had come to believe the tolls were no longer justified—the bridge having long since paid for itself. A severely split Court decided not the real, but a hypothetical, case—demonstrating once again that the framing of the issue largely determines the outcome of a controversy.

The other innovation of the Taney era came in West River Bridge Co. v. Dix (1848). Vermont had granted exclusive bridge rights. This did not prevent it from confiscating the grantees' bridge during the life of the grant— subject of course to just compensation. The Court's rationale was that all contracts are subject to higher law conditions. "Such a condition is the right of eminent domain." The indispensable power of taxation, however, is not such a "condition" and thus may be limited by state covenants, as the Court held in piqua branch of the state bank v. knoop (1854). Surely the distinction rests not on "higher law"—whatever that meant to the Court in the Piqua opinion—but upon the just compensation requirement in the one situation but not in the other, and the fact that there are many taxpayers and many ways to secure public revenue, but only one recourse when a specific piece of private property stands in the way of the public welfare.

The waite court followed a similar approach in generously defining the so-called state police power. Thus a charter to operate a lottery does not bar enforcement of a later antilottery law. "All agree that the legislature can not bargain away the police power," the Court declared in stone v. mississippi (1880). Of course neither the state in question, nor any other, has ever undertaken to "bargain away" its police power. The doctrine as enunciated in Stone is at best a truism providing no standard for judgment. Had it been used in Dartmouth, the school would have lost its case and its independence. In fact, as Gerald Gunther has remarked, the Stone police power rule has been used mainly in cases "involving prohibitions of matters widely regarded as "evil": for example, lotteries and intoxicating beverages, in an era when "Court invalidations of state laws impairing corporate charter privileges reached its highest frequency." (See inalienable police power.)

The epidemic of railroad fever that began in the Midwest in the 1850s was a prolific source of contract clause litigation. Many towns, cities, counties, and states issued railroad-aid bonds at an overall face value exceeding half a billion dollars. The purpose was to induce railroad companies to build lines convenient to the various bond issuers. Some of the desired construction never materialized. Many, perhaps all, of the railroad companies were over-capitalized. Stock watering was common. Some public and company officials were less than honest. These developments produced widespread resentment which some communities may have used for selfish purposes. In any case, what Henry Adams called the "mortgaged generation" (1865–1895) tried in one form or another to evade or repudiate much of its bond obligation. These circumstances are reflected in the path-breaking case of gelpcke v. dubuque (1864). After several decisions upholding the authority of cities to issue railroad-aid bonds, the Iowa Supreme Court, reinterpreting state law, reversed itself. On review the nation's highest Court upheld the claims of the adversely affected bondholders. In doing so it intimated that the state court's shift of position, retroactively altering the position of investors, was incompatible with contract clause principles. Popularly elected state judges apparently were more responsive to public sentiment than were appointed members of the federal Supreme Court. Ten years later Pine Grove Township v. Talcott (1874) brought the most extreme application of Gelpcke doctrine: to situations in which the state judiciary held bond issues invalid without overruling any prior decisions. The Supreme Court's rationale was that similar bond issues had been upheld in many other states before issuance of the bonds in question.

The contract clause intimations in Gelpcke and Pine Grove became an explicit basis of decision in Douglass v. Pike (1880). Much later the Court said that state court decisions did not produce contract obligations, and that neither Gelpcke nor its numerous offspring had in fact held otherwise. (Tidal Oil Co. v. Flanagan, 1924.)

Along with these bond cases another numerically important group involved the old problem of tax exemption. Without foregetting Marshall's Providence Bank rule of strict construction, the Court blocked a series of state efforts to annul pledges of corporate tax immunity.

After 1890 the contract clause as applied to state covenants gradually declined in favor or a more comprehensive, new device called substantive (economic) due process—a gross perversion of the fourteenth amendment. Years later, after the demise of that perversion, two cases suggested a possible renaissance of the contract clause. In united states trust co. v. new jersey (1977), the Court expounded a new principle of "particular" (more careful) scrutiny for cases involving public covenants "because the State's self-interest is at stake." New Jersey had issued transportation-system bonds pledging it would not substantially divert to other transportation needs the reserves and revenues securing them. Later it repealed this pledge. State courts upheld the repeal as a police power measure designed to promote additional transportation facilities. The Supreme Court reversed. No longer willing to defer to state determination of such issues, it ruled that judges must decide whether the contract impairment is "reasonable and necessary to serve an important public purpose." In this case it found that the state's needs could be served by less drastic means. In a bitter dissent Justices william j. brennan, byron r. white, and thurgood marshall insisted that for a century the "central principle" had been this: "unusual deference to [state and local] law-making authority." They could not accept a departure "from the virtually unbroken line of our cases" holding that "lawful" exercises of the police power are "paramount to private rights held under contract." The question remains, however, whether a particular exercise of power is lawful.

We turn now from public to private contract problems. No doubt the contract clause was inspired largely by debtor's relief laws, for example, measures authorizing postponed repayment of debts, installment payments, or payment in goods (often at a discount). In the Supreme Court's first encounter with this private contract problem it struck down a New York law discharging debts upon surrender of the debtor's property however inadequate to meet his obligations. Chief Justice Marshall's opinion for a unanimous Court in sturges v. crowninshield (1819) is noteworthy for two points: the "mere existence" of the national bankruptcy power does not preclude state insolvency legislation; and, while a state may not impair contract obligations, it may alter the legal sanctions (remedies, such as imprisonment for debt) for enforcement of such obligations. Eight years later ogden v. saunders (1827) construed Sturges to prohibit only retrospective application of insolvency measures. Prospective application was deemed a different matter. The Court's rationale was that a debtor relief act in existence at the time of a contract became part of the contract; later enforcement thus would not constitute an impairment. On this basis, however, a retroactive insolvency law could also be upheld. After all, a state's power to adopt future insolvency measures may equally be deemed part of every contract. In his only recorded constitutional dissent "the great Chief Justice" along with Justices gabriel duvall and Story objected: the 4–3 majority had reduced a safeguard for contract rights to no more than a prohibition on "retrospectivity." Later, we shall see, even this restriction faded away.

Marshall's dissenting view in Ogden finds support in two separate votes in the Constitutional Convention. The contract clause of the Northwest Ordinance applied only to "private contracts… previously formed." In "copying" it, the Founders (as we have seen) dropped the limiting word "private." They also dropped the limiting term "previously formed." Later, on September 15, the Convention reaffirmed this position by rejecting George Mason's motion to insert the word "previous" after the words "obligation of" in the contract clause. Thus on two occasions the convention rejected the limitation that Ogden v. Saunders read into the Constitution.

The Sturges distinction between impairment of obligations and alteration of remedies threatened in later years to undermine the contract clause. Thus bronson v. kinzie (1843), a leading Taney Court decision, taught that the allowable scope of remedial changes depends on their "reasonableness," provided "no substantial right" is impaired. As Justice benjamin n. cardozo wrote with characteristic restraint in Worthen Co. v. Kavanaugh (1935), the dividing line "is at times obscure." The leading modern case, home building and loan association v. blaisdell (1934), upheld a Minnesota "mortgage moratorium" law that extended the time of payment of mortgage loans, thus saving many homeowners, farmers, and businesses from foreclosure. The Court used the remedy and police power gambits to escape the Sturges-Ogden rule against retroactivity. The significance of the case is this: it is the culmination early in the Great Depression of a long, step-by-step process that replaced the absolute approach of the Constitution with a judicial balancing or "reasonableness" approach in private contract cases. Yet all agree these are the cases that above all else produced the unqualified language of the contract clause. Such absolutism does not mean that the founders were hard-hearted, preferring creditors to debtors. It means merely that, giving debtor-relief authority to Congress via the bankruptcy power, they opted for uniform, national treatment of the ubiquitous debtorcreditor tension.

Notwithstanding Blaisdell the "old Court" thereafter struck down several insolvency measures. One of them was louisville joint stock land bank v. radford (1935). In effect it read the contract clause into the Fifth Amendment to invalidate a federal mortgage moratorium law (a matter of reverse incorporation). Then with the advent of the Roosevelt Court in 1937 (until United States Trust in 1977) the contract clause all but vanished as a safeguard for contractual obligations—public or private. The only exceptions are Indiana ex rel. Anderson v. Brand (1938), protecting teacher tenure claims, and Wood v. Lovett (1941), protecting a tax-sale purchaser against repeal of a law that cured possible defects in his title. Wood is particularly interesting because it rests on the Fletcher principle that a state land grant entails an implied contractual obligation not to repudiate the grant in question. Then came allied structural steel co. v. spannaus (1978). There an employer had adopted an employee pension agreement. The state tried to alter it by enlarging the employer's obligations retroactively. Finding the alternation too "severe" to be upheld, the Court observed: "If the Contract Clause is to retain any meaning at all…it must be understood to impose some limits upon the power of a State to abridge existing contractual relationships, even in the exercise of its otherwise legitimate police power." As in United States Trust, Justices Brennan, White, and Marshall dissented bitterly. Stressing again their view of minimal (or no) protection of economic interests vis-à-vis the police power, they added this: the contract clause at most prevents diminution, not enlargement, of contractual obligations.

All of these cases from Fletcher to Spannaus entail a common theme: the precept of reasonable expectations. To what extent, if any, is government free to disturb those formal pledges on which men and women acting in good faith have planned their lives? A healthy legal system accommodates changing social needs. When ours was a rich and vigorous, yet underdeveloped, nation with a low overall standard of living, our law encouraged capital formation, transactional freedom, and respect for contract obligations. The "design"—not always and everywhere fully perceived—was to encourage production in the interest of a more comfortable life for everyone. Given the propensity of successful institutions to press beyond the limits of their logic, such encouragement may take grotesque forms, for example, the judicial abuses called dual federalism and substantive due process. If eventually a "backward" nation becomes highly developed, emphasis seemingly shifts from economic rights to "personal rights," from production to welfare, as in the United States beginning in the 1930s. If such a shift results in overreaction, threatening the source of the "golden eggs," emphasis may focus again on production along with protection for property and contractual rights, as in New Jersey Trust and Spannaus.

Wallace Mendelson


Hale, Robert L. 1944 The Supreme Court and the Contract Clause. Harvard Law Review 57:512–557, 621–674, 852–892.

Hurst, Willard 1956 Law and the Conditions of Freedom. Madison: University of Wisconsin Press.

Schwartz, Bernard 1980 Old Wine in Old Bottles? The Renaissance of the Contract Clause. Supreme Court Review 1979:95–121.

Wright, Benjamin F. 1938 The Contract Clause of the Constitution. Cambridge, Mass.: Harvard University Press.

Cite this article
Pick a style below, and copy the text for your bibliography.

  • MLA
  • Chicago
  • APA

"Contract Clause." Encyclopedia of the American Constitution. . 19 Sep. 2018 <>.

"Contract Clause." Encyclopedia of the American Constitution. . (September 19, 2018).

"Contract Clause." Encyclopedia of the American Constitution. . Retrieved September 19, 2018 from

Learn more about citation styles

Citation styles gives you the ability to cite reference entries and articles according to common styles from the Modern Language Association (MLA), The Chicago Manual of Style, and the American Psychological Association (APA).

Within the “Cite this article” tool, pick a style to see how all available information looks when formatted according to that style. Then, copy and paste the text into your bibliography or works cited list.

Because each style has its own formatting nuances that evolve over time and not all information is available for every reference entry or article, cannot guarantee each citation it generates. Therefore, it’s best to use citations as a starting point before checking the style against your school or publication’s requirements and the most-recent information available at these sites:

Modern Language Association

The Chicago Manual of Style

American Psychological Association

  • Most online reference entries and articles do not have page numbers. Therefore, that information is unavailable for most content. However, the date of retrieval is often important. Refer to each style’s convention regarding the best way to format page numbers and retrieval dates.
  • In addition to the MLA, Chicago, and APA styles, your school, university, publication, or institution may have its own requirements for citations. Therefore, be sure to refer to those guidelines when editing your bibliography or works cited list.