Rethinking the Nineteenth-Century Employment Contract, Again

JOHN FABIAN WITT

Legal historians have turned with renewed energy in recent years to the project of fleshing out the myriad rules by which the common law of the free labor employment contract structured social relations in nineteenth-century America. 1 Of course, labor relations have always been prominent in the literature. The German sociological tradition has long taught us to see in the legal protection of property rights a source of coercive power over the working classes. And for decades now, historians have studied the great nineteenth-century labor conspiracy cases, which generated leading cases and opinions by judges such as Shaw and Holmes. But there is a new wrinkle in recent accounts of nineteenth-century labor law. Much of the law of property, contract, and tort bears a relatively self-evident (though still too infrequently remarked on) relation to the relative bargaining power of the parties to an employment contract. Property rules, along with a whole host of attendant tort doctrines such as nuisance and trespass, allocate resources among parties. As Robert Hale observed long ago, property rules set the coercive power of A to exclude B from those resources that belong to A, whether A be a prospective employee excluding an employer from the employee's labor power, or an employer excluding a would-be employee from the means of production. In similar fashion, rules of contract and tort that define the weapons that parties may deploy in competition or bargaining also shape the relative bargaining power of social actors. Thus, doctrines of duress, fraud, unconscionability, and adequacy of consideration, and the law of labor conspiracies and competition all create immutable background rules (or sometimes inalienable entitlements) that have considerable impact on bargaining power. In Halean language, we might say that the law of duress, for example, coercively precludes the strong from forcing the weak to consent to a particular deal, or that the doctrine of fraud coercively precludes the slick from outfoxing the dupes. 2	1
The new histories of the employment relation, however, focus their attention on the development of a different subset of contract doctrines, namely those rules that judges implied into the relation between employer and employee in cases where employment contracts failed to specify particular terms. This has not been the exclusive focus of the new literature on the employment relationship. A number of recent studies have advanced our understanding of the law of labor conspiracies. But the distinctive feature of the new studies is their elaboration of the incidents of the employment relation. In particular, the new histories of the employment contract contend that the nineteenth-century law of employment constructed a prescriptive status hierarchy through the judicial elaboration of implied doctrines of contractual construction.	2
This essay argues that analysis of the social consequences of these implied, judge-made rules involves an intricate problem of interpretation that the new historical literature has not addressed. The rules in question are often default rules rather than immutable rules. 3 Default rules govern employment contracts only in the absence of some contrary indication of the parties' intent, and they are fully amendable in any particular transaction. As a result, default rules of contract interpretation do not have any straightforward impact on the relative bargaining power of the parties. Unlike property rules, the default terms of the employment contract are unrealizable absent entry into a contractual relation; 4 and unlike immutable contract rules (such as the rule against contracts of self-enslavement, or the doctrines of fraud and duress), default terms are not binding elements of an agreement. In other words, default terms confer neither a valuable entitlement nor an inalienable duty on either party to a bargain. Moreover, they do not (at the level of formal analysis) have any sticking power. Unlike the law of the nineteenth-century marriage relation, for example, which set the immutable terms of the legal relation of husband and wife, the default terms of the nineteenth-century employment relation did not prescribe a particular mandatory regime of authority and subordination in the workplace. A party with the foresight to consider the eventuality of a work accident, the denouement of a particular work relation, or the relations of daily control over work processes could have demanded a particular arrangement, and the fact that the default term was one way or the other did not—again, at the formal level of analysis—change the chances that the party would be able to get what she wanted.	3
The distinction between default rules and immutable rules requires a rethinking of this strand of the new histories of the nineteenth-century employment contract. Part 1 of this essay describes the new histories in more detail. Part 2 turns to the social practice of contracting out of the default terms of the employment contract, which (though limited) was considerably more widespread than the recent literature suggests. Part 3 outlines the limits of the new histories' only explanation of the social consequences of default rules, namely that they had an ideological or norm-shaping effect.	4
This is not to say, of course, that the default terms of the law of employment did not matter. The point here is that the new histories of the nineteenth-century employment contract do not have an adequate account of why they mattered. Part 4 sketches the beginnings of an alternate account of the social consequences of the common law of employment. In particular, it suggests that the technical complexity of the default rules of the employment contract, along with the daunting intricacy and unpredictability of the rules governing what constituted sufficient and cognizable evidence of contracting out of those defaults, may have had greater social consequences for the subordination of employees to employers than the substance of the default rules themselves.	5
A word at the outset about the ambitions of this essay may allay confusion. I do not seek to establish a wholesale revision in thinking about the legal construction of the employment relation. It is my view—and I take this to be a quite traditional view—that the rules described above as Halean were the central players in the construction of the nineteenth-century free labor employment contract. Rules that allocated resources, whether by setting property allocations or (as in the case of the law of labor conspiracies) by limiting the capacity of particular groups to augment their bargaining power, determined what employees could get from employers and what employers could get from employees. 5 Moreover, such legal rules appear to have had powerful ideological effects on American labor relations; over time the nineteenth-century law of labor conspiracies drained a good deal of energy from the organizing of American workers, making still more difficult what little collective action the common law allowed. 6 This essay, however, is not about the effects of Halean rules on the employment contract. It does not consider such issues as the law of labor conspiracies and unequal property allocations between employers and employees. Instead, it focuses on the social consequences of default rules in the employment contract. If I am successful here, I will have convinced you only that on this relatively narrow, but nonetheless important, question, the historical problem is considerably more complicated than our recent accounts indicate.	6
I. Common Law Default Rules in the Employment Contract
A central contention of the new histories of the law of free labor is that the nineteenth-century law of the employment contract assigned the vast majority of American employees to the subordinate position in a workplace status hierarchy. The claim is not merely that legal coercion characterized the employment contracts of former slaves and the very poor under the Black Codes, the Freedmen's Bureau, and the vagrancy laws; nor is it the neo-Halean view that the law of property, contract, and tort allocated resources in such a way as to systematically disadvantage workers. 7 Rather, it is that the law of the labor contract itself constructed a prescriptive status regime. In the words of Christopher Tomlins, autonomy in labor relations was constrained "not just by burgeoning material inequalities [i.e., the law of property] but by doctrinal constructions of the social roles interpolating legal conditions of existence." In Tomlins's view, the law of the labor contract "showed itself not just as a framework for the realization of private power and domination but in fact as an element integral to their construction." Stated more succinctly by Karen Orren, the new critical view of the labor contract posits that nineteenth-century courts "continued to prescribe labor relations, not derive them from existing contracts devised by the parties." 8	7
In particular, four legal doctrines form the core of the argument that the law of the employment contract constructed a relation of prescriptive status: the "entire contract" doctrine; the law of enticement; the assumption of employer control; and the law of workplace accidents.	8
The "Entire Contract" Rule
Commercial law has to answer the question of what to do when a contract for services is only partially performed. Early nineteenth-century American courts faced this problem in the context of building contracts and labor contracts. In the building contract context, the leading case of Hayward v. Leonard, decided in 1828 by the Massachusetts Supreme Judicial Court, held that commercial builders could recover in quantum meruit for services rendered in partial performance of contract requirements. 9 But in labor contract cases, the new labor law histories contend that courts applied a different rule. Under the "entire contract" rule, the breaching employee who had served only a portion of the term of her labor contract could not recover back wages for work actually performed. 10	9
Several historians have taken issue with this characterization of the "entire contract" doctrine. 11 Indeed, the case law appears to have been varied and unpredictable. (More on the effects of rule complexity later.) Over the course of the nineteenth century, American courts increasingly allowed workers to recover back wages in quantum meruit actions. By the end of the nineteenth century, American jurisdictions were split about half and half between a rule that allowed quantum meruit actions and a rule that denied recovery to workers who quit.	10
Enticement
The "enticement" doctrine provided employers with a nonreciprocal right to sue for tortious interference with the employment contract. Employers could bring an action for damages against a party who interfered with their employees' performance, but employees rarely had the reciprocal power to bring such an action against parties who interfered with the fulfillment of employers' contractual obligations to their employees. 12 Moreover, the nineteenth-century law of enticement allowed tortious interference claims by employers even where the employment relation was on an at-will basis rather than for a term. 13	11
The Assumption of Employer Control
Both the entire contract doctrine and the enticement rule effectively promoted employers' control over their workforce; the former rule created obstacles to quitting before the end of the contract term, while the latter decreased the competition for workers' services. The nineteenth-century law of employment, however, also developed explicit doctrines of employer control. As Lemuel Shaw ruled in the 1833 case of Sproul v. Hemmingway, an employment relation existed between parties if one was subject to the "order, control, and direction" of the other. 14 Thus when an employer purchased a worker's labor time, the employer bought the "[f]aithful service" of the worker in the pursuit of all "lawful and reasonable commands." 15	12
Workplace Accidents
The nineteenth-century law of work accidents notoriously denied workers recovery from their employers for injuries arising in the ordinary course of employment. 16 Beginning with the case of Farwell v. Boston & Worcester Railroad, American courts held that employees assumed the risk of injuries arising from the "natural and ordinary risks" of employment. Thus, employees could not recover for injuries arising out of the negligence of their fellow servants, even when the negligent fellow servant had been in a position of superintendence over the injured employee, or when the negligent fellow servant had been employed in a different department of the firm, beyond the influence of the injured employee. 17	13
Over the course of the nineteenth century, the harsh work accident rules eroded considerably; by the end of the century, the law of work accidents had become, in the words of one New York lawyer, a "hodge podge" of inconsistent and contradictory rules and standards. 18 Still, few workers injured in the workplace were able to recover in tort actions against their employers until the enactment of workmen's compensation statutes in the 1910s. 19	14
II. Contracting Out of the Common Law Defaults
Together, these common law doctrines of the employment contract created a formidable body of legal principles that systematically favored employers in litigation with employees. But unlike immutable rules of private law such as the doctrines of duress and fraud, or the law of collective employee action, each of these doctrines set default rules that applied to a given employment contract only if the parties did not make an agreement to the contrary. This has important consequences for measuring the social consequences of legal rules. Absent some impediment to contracting, the parties to a contract under a default regime may allocate rights and responsibilities in their contractual relationship without regard to the background rules of contract law. 20 Moreover, at first glance there is good reason to think that the default settings of the law of contracts will have either no impact or (at best) a highly contingent impact on the relative distribution of wealth between the parties to a transaction. 21 Contract defaults, after all, do not assign an entitlement to anyone, because no one party can realize the benefits of the default absent a relationship with some other party. To be sure, an employer may be able to coerce (or bribe) a less powerful employee into accepting defaults favoring the employer. But if the default rule favors the employee, that same employer—and this is the critical point—would be able to use that same market power to coerce (or bribe) the employee into switching to a pro-employer term. If so, it is hard to see what work the initial setting of the default rule performs. 22	15
One need not assume that human beings are autonomous, self-seeking, and rational actors, or that (as Judge Posner has recently written) human behavior is analytically analogous to the behavior of rats, 23 in order to grant that the ability of employers and employees to contract around default settings poses a significant analytic complication for the new histories of the employment contract. In fact, the historical record indicates that many employers and employees contracted out of the default terms of the law of the labor relation.	16
A number of mid to late nineteenth-century American firms, for example, contracted around the rule of employer control through "inside contracting." Under this system, skilled workers in industries such as iron production controlled the management of the production process, contracting with the firm's owners only for the total tonnage of iron to be produced and the tonnage rate. 24 Similarly, across the second half of the nineteenth century, trades unions of skilled industrial craftsmen successfully set union rules for the management of the workplace, requiring that employers hire only union members and setting wages, hours, and conditions of labor. 25 Even unskilled workers effectively contracted around the common law rule of employer control when they opted for piecework labor that they could do from home, outside the supervision of their employers.	17
Other employers and employees contracted out of the common law of work accidents. Some trade unions, for example, developed a sophisticated Marxian theory of contracting around the default rule of liability for work accidents by exacting higher wages in order to fund trade union insurance funds. Marx argued that wages were determined by the subsistence requirements of the worker. Those subsistence requirements, in turn, were "themselves products of history," and depended "on the conditions in which, and consequently on the habits and expectations with which, the class of free workers has been formed." 26 According to the German-language labor newspaper New-Yorker Gewerkschafts-Zeitung, it followed from the Marxian theory of wages that payments into a union relief fund would, over time, heighten workers' expectations and thus result in higher pay scales. At the very least, the increased union solidarity fostered by the relief funds would increase the union's bargaining power and thus increase wages. "So," the editors concluded (a bit too optimistically, no doubt), "it is already in our power to make the employers financially liable for the care of their victims." 27	18
On the employers' side, a small minority of firms created employee accident funds in the last years of the century to which employer and employee made regular contributions for the compensation of accident victims. The creation of such funds was limited to a few large firms—frequently railroads, where the problem of work accidents was particularly acute. 28 Moreover, courts limited the enforceability of the waivers of the employee's right to sue that were a part of the work accident funds. 29 But in those firms that did establish a work accident compensation plan, it became a part of the standard employment contract.	19
Indeed, employment contracts crafted novel arrangements across a wide variety of employment issues. Employers and employees, for example, contracted into their own arrangements for the rules governing ownership of the intellectual property rights in employee inventions. 30 Employers also readily used their market power to contract out of default contract terms in cases in which state legislatures readjusted the background term. Most famously, employers showed little regard for the eight-hours laws that were enacted in a series of states in 1867 and 1868. 31 The first statute making eight hours a legal day's work was enacted in Illinois in the spring of 1867. But like the statutes that followed, the Illinois statute allowed the parties to bargain to an alternative agreement; as a result, employers such as the McCormick reaper works simply maintained their ten-hour day. Similar results accompanied the Massachusetts eight-hour statute of that same year, and even the eight-hour rules on federal public works projects. In each instance, and especially in Illinois, employers' attempts to contract around the new eight-hour day were met by fierce resistance among workers, who walked off the job after eight hours and struck to make the new default rules stick. The strikes were failures, however, and workers in Illinois, Massachusetts, and on federal public works projects quickly found themselves back on the job under their old terms. In the case of the hours of labor in a legal work day, then, the content of the default rule appears to have been of little import. 32 Instead, what mattered was the legal enforcement of property rights and the host of legal rules of contract and tort that allocated bargaining power among employers and employees, such as the law of labor combinations, duress, and fraud.	20
This, at any rate, was the lesson that many American labor radicals took away from the debacle of the eight hours laws. George E. McNeill led the eight hours movement in Massachusetts as secretary of the Grand Eight Hour League and then as president of both the Workingmen's Institute and the Boston Eight Hour League. 33 In 1883, he joined the Knights of Labor, and throughout the 1890s he maintained close ties to the AFL. According to McNeill, the wage labor system usurped the independence of the free laborer and transformed him into "a man without the rights of manhood." 34 But after his experience with the eight hours enactment, McNeill abandoned efforts at effecting legislative change through amendment of the background rules of the employment contract. Instead, in the late 1870s McNeill turned to the purely economic action of organizing workers into unions so as to increase their bargaining power vis-à-vis their employers. 35 Similarly, Eugene V. Debs appears to have appreciated the significance of the default/immutable distinction. Debs's first proposal as a state representative in Indiana in 1885 was to amend the law governing railway employers' liability for injuries to workers. For Debs, however, it was critical that the law of employers' liability bar railway employers from contracting out, and when the state senate stripped his bill of the bar on contracting out, Debs refused to vote for it. Debs, in other words, insisted that employers' liability rules be immutable rules rather than default rules. 36	21
Of course, it is one thing to say that some employers and employees contracted around the default rules of the employment relation, or that certain labor leaders understood the futility of switching default rules without either augmenting workers' bargaining power or limiting employers' ability to contract around the defaults. It is altogether another to say whether courts upheld or defeated such attempts at negotiating novel arrangements. For example, a number of courts (especially late in the nineteenth century) were reluctant to uphold employment contracts in which employees wholly assumed the risk of injury at work by waiving the right of the employee to bring his common law action against the employer. 37	22
Indeed, a wide range of principles of contract interpretation shaped and constrained the capacity of the parties to bargain around default terms. In part this was the product of the very real difficulty of determining when there was a gap in a contract's express provisions that needed to be filled with a legal default rule. 38 But it may also have been the result of background rules of contract interpretation. For one thing, the development of a regime of default terms may lead courts to imply the default term rather than do the work of interpreting ambiguous or contested provisions. 39 Even more important, a host of contract law doctrines created substantial obstacles to successfully contracting into innovative employment arrangements. The "clear statement" rule required unambiguous expressions of intent to contract out; the parol evidence rule barred consideration of oral promises made outside an integrated writing; the fuzzy distinction between completely and partially integrated agreements and the collateral agreement exception to the parol evidence rule determined when parol evidence could be entertained; the plain meaning rule barred extrinsic evidence to clarify the meaning of a document clear and unambiguous within its four corners; and the one-year provision of the Statute of Frauds required that long-term employment commitments be made in writing. All of these doctrines made it difficult for employees to establish that particular employment contracts had, in fact, sought to establish conditions other than the defaults. The extent to which judges allowed employees to testify to promises and conditions outside the text of the posted work rules, for example, would have played a critical role in determining employees' capacity to contract into alternate employment conditions.	23
III. Accounting for the Effects of Default Rules: The Norm-Shaping Hypothesis
Ultimately, it appears that relatively few employers and employees contracted around the default settings in the employment contract. The work arrangements of the skilled industrial craftsmen who contracted for almost plenary control over the workplace never characterized more than a small proportion of American labor contracts. Likewise, few workers were protected by employer accident funds, and few employees were concerned with the development of new inventions on the job. Even fewer workers appear to have negotiated express rights to back pay for work actually performed in the event of the employment relation ending prior to the conclusion of a term, or to have negotiated the right to be approached by outside employers. The default terms of employment law, then, appear to have constituted the terms and conditions of the labor relation in the majority of American free labor work relationships. Employers and employees could and did contract around them. Yet by and large, they simply stuck with the default terms implied by law.	24
The student of nineteenth-century employment law thus confronts a field of what Ian Ayres and Robert Gertner have usefully called "sticky" defaults. 40 Alternatively, it might be said that the default rules of the nineteenth-century employment contract appear to have exerted a gravitational pull on the shape of the employment relation. The mechanism by which default terms developed their sticking power, however, remains deeply obscure.	25
Standard economic theories of law have trouble explaining the gravitational pull of the defaults. Law and economics approaches have traditionally interpreted the failure of parties to contract out of default rules as a measure of the rules' efficiency. 41 But this can hardly be right in the employment context. Conditions of employment varied widely in nineteenth-century America, from small handicraft shops to giant railroads, iron production behemoths, and huge mines. If employment contracts were efficiently gauged to particular workplaces, we would expect them to exhibit a similar diversity. 42 Nor do transaction costs—the other standard law and economics explanation—offer a persuasive account of default term stickiness in nineteenth-century employment law. The transaction costs of reallocating default terms are not—and were not in the nineteenth century—sufficient to account for the staying power of the defaults. It was cheap and easy for employers to adopt new terms in employment contracts. Nineteenth-century employers created binding work rules that were incorporated into the employment contracts simply by posting them where workers could readily see them. 43 Employees, by contrast, may have faced considerably higher transaction costs in seeking to amend the terms of the employment contract. But given how inexpensive it was for employers to offer alternatives to the default rules, law and economics approaches would expect competition for labor among employers (which was often quite intense during much of the nineteenth-century, especially for skilled labor) to generate differences in the employment packages offered by different employers. 44	26
The account of the common law default rules offered by recent histories of the employment contract is unsatisfactory as well. As we have seen in Part 2, it is not quite accurate to contend that the law of master and servant necessarily assigned employers and employees to particular positions in a status hierarchy. Some employers and employees did contract out of the defaults—indeed, contracting around defaults occurred more often than recent accounts would suggest. 45	27
It should hardly be surprising that our histories of the employment relation do not yet have a good account of the role of default terms. Marx and Weber, who set the framework for much of our analysis of the employment relation, themselves failed to analyze the distinctive features of default rules. Both were more interested in the effects of disparate resource allocations. Marx's "reserve industrial army" of the unemployed was subject to the "silent compulsion of economic relations," in large part because it had been forcibly freed of the means of production in the early modern expropriation of the peasant from the land. For similar reasons of unequal resources, Weber thought that formal juridical equality gave rise to ever-increasing relations of authority and subordination. 46	28
Unlike Marx, Weber at least noted the distinction between permissive (or default) rules (ius dispositivum) and mandatory rules (ius cogens). Default rules, Weber explained, served the convenience of parties to contractual arrangements who through "considerations of mere expediency" failed to account for every possible contingency in their contracts. "As a rule, the parties do not think of really taking care of all the possible relevant points." In passing, however, Weber did suggest one possible additional role for ius dispositivum. Their "even more fundamental significance," he noted, might be that they exerted a kind of "normative control" over the parties. 47	29
Alas, Weber never elaborated on the normative import of the default/mandatory distinction. It is this possibility of defaults shaping social norms, however, that Christopher Tomlins has pursued as an answer to the default rule conundrum. The general thrust of his argument often elides the distinction between default rules and immutable rules. 48 But Tomlins's ultimate contention is that the common law of the employment contract had an ideological and normative power in the creation of American work relations. Labor law, Tomlins argues, "authorize[d]" the authority and subordination of employer and employee and "inscribe[d]" that relationship on individuals. Employment law, in his view, had the power to name and thus construct social relations, and it put its powerful stamp of approval on relations of domination "by declaring legitimate subordinations, of the not-free, [or] not-master." 49	30
The neo-Weberian norm-shaping argument has many virtues. Ultimately, it must be considered in any explanation of the relationship between legal default rules and labor relations, especially during times of historical change and flux. But the assertion that law enjoyed a "preponderance of ... authority" 50 as a source of norms is extremely difficult to substantiate. There are, of course, in any culture competing sources of cultural norms, whether they be found in politics or in religion or in the arts or cultural life more broadly. (Hendrik Hartog's essays on the often disjointed relation between formal legal rules, on the one hand, and social practices, on the other, make this point quite eloquently for several nineteenth-century fields outside of employment law.) 51 And in nineteenth-century America, many extralegal sources of norm creation such as artisanal republicanism and producerist ideologies fostered ideas of freedom and self-direction in the workplace. 52	31
Moreover, it is often difficult to describe in general terms the kinds of social norms that any particular authority will generate. The social impact of a common law rule is highly contingent on the structure and idiosyncracies of the particular social or market context in which the rule is implemented. 53 It would, in fact, be surprising if it were otherwise given the competing strands of meaning immanent in most cultural production. The ideology of free labor, for example, gave rise to contradictory norms about the organization of work: If employers seized on it to rationalize their own power in the workplace, employees appropriated it to argue for worker control, independence, and dignity. 54 Indeed, I have argued elsewhere that the common law of work accidents is best understood as containing within itself precisely this kind of ambiguity. The outcomes of particular cases were frequently inhospitable to workers who had been injured. But at the same time the work accident cases reaffirmed a constellation of mid-nineteenth-century ideas about workers' authority and control over the conditions of the workplace. Niggardly work accident doctrines may even have structured financial incentives such that employers were not pressured to encroach on cultures of worker control in order to reduce work accident costs, if only because those costs were negligible. When workmen's compensation legislation replaced the old common law rules, however, American employers responded to the increased cost of work accidents with hierarchically organized managerial structures. 55	32
At a deeper level, the Tomlins argument does not provide a fully developed theory of the relationship between the law and individuals' beliefs. The argument requires that, at some level, individuals come to believe that workplace relations are inegalitarian because the law says so. But the mechanism by which the law's norms come to be absorbed into belief is extremely difficult to discern. Who is it, for example, who comes to believe law's norms? Employers? Consumers? The voting public? Or the workers themselves? And how?	33
In this regard, the neo-Weberian theory of norm legitimation offered by Tomlins should be distinguished from other recent theories of law's ideological effects. Duncan Kennedy has observed that the distinctive claim of recent historical approaches to the employment relation is that the law often not only constructs the relative bargaining power of the parties, but also shapes the worldview of the people that it regulates. 56 This observation is surely correct, and the constitutive role of law has made important contributions to our understanding of the relationship between law and society. 57 Yet it must also be said that some theories of norm legitimation in the law are more plausible than others. Consider, for example, the highly plausible Gramscian theory of norm legitimation. On the Gramscian account, legal regimes allow certain kinds of rights claims. But those claims, in turn, legitimate or reproduce the premises of the legal regime. Rights claimants may attain a certain degree of autonomy, or perhaps certain resources, but only at the cost of reproducing the social order on which their claims of right rest. Thus Eugene Genovese's law of slavery at once constrained and legitimated the power of Southern slaveowners by allowing certain slave claims of right, or by requiring that slaves make such claims on their masters rather than on the state. 58 Similarly, in recent accounts by Ariela Dubler, Katherine Franke, and Hendrik Hartog, the late-eighteenth and early-nineteenth-century law of marriage allowed women to make certain claims of right in their marriages, but only by legitimating the incidents of the marriage relation generally. 59	34
Of course, the Gramscian approach to norm legitimation is often hotly contested. Slaves' and married women's rights claims frequently represented strategic exploitation of loopholes in their legally prescribed status rather than constitutive reproductions of that status. The point here is that unlike the Gramscian theory, the Tomlins account of legitimation faces real problems in explaining how the norms implicit in legal rules came to be social norms. This is not to say that the norm legitimation argument has no place in an account of the social consequences of default rules. Even if the default rule has normative power only at the margins, after all, it would affect the labor market. 60 But why should we think that workers followed as children to the law's Pied Piper when the Piper told them that they were subordinate to their employers? In addition, when applied to the public's ideas about employment, the norm-shaping argument has a good deal of difficulty insofar as it argues from judicial rhetoric rather than adjudicative outcomes. 61 There is good reason to doubt the power of the dicta and reasoning of appellate decisions to shape public opinion. 62 Indeed, to the extent judicial opinions on controverted issues make their way into the public eye, there is little reason to attribute more norm-shaping power to nineteenth-century employment law decisions than liberals now attribute to the decisions of Justice Thomas, or conservatives attributed to the opinions of Justices Brennan or Douglas.	35
IV. Retheorizing the Social Consequences of Default Terms
The critique offered here has not so much advanced a new theory of how to conceptualize the social consequences of common law default rules as it has sought to suggest that such a new theory (or set of theories) is necessary to make sense of one important strand of the new histories of employment law. Indeed, this essay is hardly the place for the development of a full-blown account of the ways in which defaults shaped the employment relationship—in part because the social consequences of contract defaults are highly contingent on the specifics of particular circumstances. Yet in this final section I want to advance some speculations on ways in which legal historians might begin to investigate the meaning of contract defaults in the employment relation.	36
Some Possible Accounts of the Social Consequences of Common Law Defaults from the Contracts Literature
One approach to the social consequences of default rules might be located in the new literature of the behavioral sciences. 63 In a sense, this literature merely confirms the power of inertia in human behavior. When people are endowed arbitrarily with a particular entitlement, they tend to value it more highly than people not endowed with the same entitlement—the so-called "endowment effect." Default rules, of course, do not really endow parties with any entitlement at all because they come into play only in the event of a relationship between parties. Yet there is evidence to suggest that by creating a template for agreements, defaults create a perceived baseline departures from which are seen as deviations from a status quo ante. Parties to contract negotiations appear to require their negotiating partners to pay for switching a default rule—and the partners appear to be willing to pay. 64 Thus, the default settings in the labor contract may have led employers to value limited accident liability, workplace control, protection from enticement, and the entire contract rule, more highly than employees. If the default settings had been different, employees may well have been able to exact higher wages for allowing employers to bargain for such terms. Alternatively, employees might have sought to retain workplace control, to impose accident liability on employers, to retain the right to entertain offers from outside employers, and to retain the right to back wages in the event they quit or were fired before the end of the full contract term. In this case, the endowment effect suggests that under pro-employee defaults, employees may have been able to retain favorable terms for themselves at lower cost in decreased wages.	37
Another strand of behavioral theory offers a possible explanation for the stickiness of nineteenth-century default terms. "Prospect theory" posits that individuals weigh the negative consequences of action more heavily than the negative consequences of inaction. In other words, they experience greater "regret" costs when departures from the status quo turn out badly than they do when the status quo itself turns out badly. All other things being equal, therefore, individuals will prefer to maintain the status quo than to depart from it. (In part, at least, this is because individuals perceive the reputational costs of unsuccessful innovation as higher than the reputational costs of failure arising out of behavior consistent with conventional wisdom.) 65	38
It follows from the regret theory of defaults that employer and employee preferences for contractual terms may have been endogenous to the default settings. Employers, the theory suggests, may have become attached to default rules that inured to their benefit and thus reluctant to trade them away for cash (lower wages). Employees, too, may have become reluctant to trade away cash in order to purchase terms such as accident liability. Thus, the regret theory of defaults potentially offers a strong explanation for the stickiness of default rules. Employers and employees would not have bargained around default terms, even when we might have expected them to, because of psychological attachments to the rules as they were given.	39
The difficulty with behavioral science explanations of nineteenth-century employment practices, however, is that theories of status quo bias have considerable trouble dealing with historical change. How, after all, do individuals develop attachments to a particular state of affairs in times of institutional flux? The employment contract was changing dramatically in the early and mid-nineteenth century. Craft employees were increasingly unlikely to live in the homes of master craftsmen across the 1810s and 1820s. Large manufacturing firms were being developed for the first time in the United States in the 1820s and 1830s. Indeed, by the 1870s and 1880s the organization of employment in certain large industrial firms was hardly recognizable from the perspective of just a few decades earlier. At the same time, the law of employment was going through a series of important changes. The law of master and servant, with its default presumptions of employer control, only came to encompass all contracts of hired labor in the first half the century. The law of work accidents was an invention of the late 1830s and early 1840s. In the Reconstruction South, employers, employees, lawyers, and judges were required to create new rules of free labor from whole cloth. And it was only after the publication of Horace Wood's 1877 treatise on master and servant that the at-will rule became the default position for employment contracts. 66	40
Indeed, underlying the problem that instability and change pose for the idea of a status quo bias is the deeper problem of determining the level of generality or abstraction at which people will fix any given idea of the status quo. The relevant baseline for measuring the status quo in the nineteenth-century employment context may not have been any one particular arrangement of duties and rights. Rather, the status quo may have been a higher order condition of flux, of change and departure from prior practices. In a situation of continual rearrangement of work relations, then, it might be reasonable to expect employers and employees to have contracted out of default terms with considerable frequency, whenever doing so suited their needs.	41
A more helpful approach to understanding the social consequences of particular default rules might follow Ian Ayres and Robert Gertner's observations about the informational function of contract defaults. Ayres and Gertner argue that the failure of a contract to specify a term for a particular contingency (thus incorporating by default the common law's implied rule) may result from opportunistic bargaining by parties with asymmetrical information. Proposing a particular term often has the effect of imparting otherwise private and valuable information to the other side. Thus, if a party with superior information is required to contract around a default term, that party will thereby be required to reveal otherwise withheld information. By contrast, a default rule that allows parties with superior access to information to remain silent facilitates opportunism and exploitation of information asymmetries. 67	42
The employment contract presents an obvious target for such strategic exploitation of information asymmetries. Employers are likely to possess an abundance of information about working conditions, about the financial soundness of the firm, and (as repeat players in employment contracting) about the nature of the default rules of employment law. This puts them in a particularly good position to achieve informational advantages and to utilize them to the detriment of prospective employees. Consider, for example, the default rules for workplace accidents. Employers in the nineteenth century were in a much better position than prospective employees to possess information about the risk of work accidents. Employers knew, in an approximate way at the very least, how many of their workers had been injured on the job. 68 Employees and prospective employees, on the other hand, may have heard anecdotal reports from family, friends, and coworkers. But they did not have easy access to the kind of global work accident information that employers had (a problem that was exacerbated significantly by the legal restrictions imposed on union organizing). Moreover, in the most dangerous industries, employers were likely to have detailed knowledge of the default rules that, for the most part, insulated them from liability for the work-related injuries of their employees. By giving the benefit of the default to employers, the law of work accidents facilitated the exploitation of informational advantages in the employment contract. Employers with high work accident rates could remain silent and thus not tip their hands by trying to contract out of liability for work accident injuries.	43
The Consequences of Rule Complexity
Of these possible explanations for the social effects of the common law defaults, the Ayres and Gertner information argument holds the most promise for future explorations of the historical consequences of employment law defaults. Yet it seems unlikely that the aggravation of asymmetrical information problems completely captures the impact of common law defaults in the employment law context. In this final section, I suggest that the consequences of the law of employment may not have been a result of the substantive content of the common law rules but rather a product of the technical complexity of their articulation and the institutional settings in which they came into play between employer and employee.	44
As a threshold point we might want to begin by saying that the policy choice to resort to a contract law regime at all, and in this case a contract law system in which the parties were allowed to opt into varying terms and conditions, was significant. Domestic partners, after all, were not allowed to select their own terms. 69 And the South offered readily available examples, both before and after the war, of varying systems of labor that employed compulsory contracts and compulsory terms within those contracts. Yet as a realistic alternative, it is difficult to imagine a mandatory regime of publicly prescribed terms of labor gaining a foothold in the Northern free labor economy. As David Montgomery explained some time ago, even labor radicals in the immediate post-Civil War years were unwilling to part with the contractarian principle that individual parties ought to be free to negotiate their own deals in employment. 70	45
Even within a regime of formal freedom in the negotiation of employment relations, there were significantly different background approaches to the law of contracts that might have had widely varying social consequences. Of particular interest here is the persistent adoption by nineteenth-century courts of default rules rather than default standards in the adjudication of employment cases. Rather than taking a case-by-case contextual approach to employment cases that might have sought to tailor gap-filling default terms to the particular employment relations at issue, courts laid down rules for application across a wide range of employment cases. The fellow servant rule and the doctrine of assumption of risk, for example, were rules that set up threshold obstacles to underlying standards. They stood as preliminary hurdles to the general negligence standard of nineteenth-century tort law. Yet Lemuel Shaw might simply have announced that the default term controlling which party bore the cost of workplace injuries should be the term that would best reduce the number of accidents. This reasoning would thus have represented a general standard for courts to apply in subsequent cases. (Imagine Shaw ruling that "The cost of work accidents is chargeable to the party best able to reduce the rate and severity of accidents.") Instead, he laid down a rule that (implausibly) viewed the employee as the best cost avoider across all working conditions.	46
By adopting a regime of rules rather than a regime of standards, the historian might argue that courts advanced the interests of employers over those of employees. 71 For one thing, the choice of rules over standards significantly reduced the discretion of juries and increased the power of the judiciary at both the trial and appellate levels. Furthermore, it was the employer who was a repeat player in the employment contract market and who was better able to structure its negotiations around a baseline of finely articulated legal rules. The employee, by contrast, was likely to be a less experienced player in the labor contract market, facing a complex skein of legal rules. Employers were also likely to have considerably more litigation experience and thus to know better the precise legal import of particular terms and conditions. Repeat appearance in the market and the courtroom probably provided many employers strategic, specializing, and institutional advantages in the legal arrangement of the employment relation. 72	47
Yet as an account of default rules, the rules/standards dichotomy may be less helpful than it seems at first glance. With respect to the parties' capacity to contract out of the defaults and set their own terms, the choice between rules and standards may matter very little. There is even good reason to think that the choice of a relatively vague standard may induce parties to contract into more precise arrangements of their own, particularly in the case of high-likelihood contingencies. 73 Moreover, rules are frequently standards in disguise. 74 And in many instances, the rule/standard dichotomy appears to break down in the nineteenth-century law of employment. The law of workplace accidents quickly became choked with exceptions and counterrules, recourse to which undermined the rule-like certainty of the fellow servant rule. By the end of the nineteenth century, the law of work accidents had become so extraordinarily complicated that the "Master and Servant" section of the Century Edition of the American Digest dedicated 662 of its 1,011 pages, and 1,073 of its 1,289 headings to the rules of employer liability for work accidents. 75 Similarly, the varied and unpredictable outcomes of cases decided under the entire contract doctrine suggest that, as in Carol Rose's description of nineteenth-century property law, crystalline rules gave way to muddy, standardlike concerns such as the good faith of the parties. 76 And indeed, in employment litigation equitable doctrines such as unjust enrichment, quantum meruit, waiver, and abandonment for cause frequently undermined the ex ante certainty of hard rules. 77	48
In fact, the picture of nineteenth-century contract law that emerges from the most exhaustive surveys of the cases is one of a highly complex and unpredictable system of background defaults. Lawrence Friedman's study of employment contract cases in Wisconsin, Peter Karsten's uneven but still useful survey of appellate cases, and Louise Wolcher's investigation of duty to mitigate decisions all highlight the lack of consistency and the flexibility of doctrines such as fraud, misrepresentation, and waiver. Outside the leading cases in which great judges issued ringing pronouncements in favor of particular rules, run-of-the-mill litigation seems to have been considerably less consistent in result than the elaborated doctrinal rule schemes might suggest. 78	49
This does not mean, to be sure, that the law of employment was somehow employee-friendly. Quite the opposite. The complexity and unpredictability of judicial approaches to employment law cases may well be critical to understanding the infrequency with which employment contracts changed the underlying default rules. At the very least, the rule complexity of the law of employment appears to have made it prohibitively expensive for the overwhelming majority of wage laborers to litigate disputes with employers. Friedman's contract study found that even though labor and service contracts represented a leading category of cases, most occupations in American life were wholly unrepresented in Wisconsin employment cases. Willard Hurst's study of employment contract decisions in that same state's logging industry makes the same finding for work accident cases. (Indeed, outside the railroad context, few American workers appear to have brought work accident cases in the nineteenth-century.) And more recent investigations of employment cases show the dockets filled with cases involving managerial and supervisory employees rather than the run-of-the-mill workers. 79 Almost all of these studies, of course, rely on reported appellate cases rather than trial court records. Yet the power of employers to appeal adverse trial decisions must have been a major deterrent to employee litigation over small claims. The very fact of the law's complexity and unpredictability provided extra leverage to the party with the resources to appeal adverse decisions.	50
Even more significant for our purposes, a background default regime of complex and unpredictable rules probably helped to diminish contracting out by employees. For employers who were able to spread the cost of legal expertise and contract negotiation across multiple employment contracts, unpredictability in a default term may have prompted contracting into particular terms with greater certainty. 80 But where the background defaults were not just unknown, but were also highly unpredictable and technically daunting, employees had less reason and ability to know that they preferred a different term. Compare, for example, an employee in a system of simple and clear employee-unfriendly defaults with an employee in a system of complex, unpredictable defaults. The latter employee would have weighed the not-inconsiderable costs of negotiation against the cost of sticking with defaults that may or may not have been against him. But the cost of inertia would have to be discounted for the unpredictability of the outcome. The former employee, on the other hand, would have weighed those same negotiation costs against a starkly unfriendly default regime that would not have been discounted for the possibility of its turning out to be less unfriendly than it appeared.	51
Moreover, successfully contracting out was itself a complex legal maneuver for employees, requiring an understanding of precisely which talismanic phrases would effectively convey the intended legal construction. Consider, for example, the plight of Bridget Daveny, a domestic servant in New York City in the late 1870s. Daveny sought to contract into a term of service through the entire summer, rather than the month-to-month service default rule for domestic employment. "You keep your help for the summer months?" Daveny asked her employer, who replied, "Oh certainly, and longer if they suit me." At the end of a one-week trial period, the employer told Daveny "You suit me if I suit you." Daveny responded: "As long as I suit you, there is no fear for the summer months," and her employer replied, "Oh, yes, I like you very much." After two months, however, Daveny's employer told her that her services were no longer needed. And despite Daveny's apparent attempt to make abundantly clear her understanding of the employment contract, an appellate court focused on the conditional nature of the parties' language to hold that the oral employment contract had not contracted out of the default rule of a domestic hiring from month to month. 81	52
Daveny was hardly alone in failing to make her deal stick. Nineteenth-century reports are littered with cases of employees who, for technical reasons such as the parol evidence rule, failed to exact enforceable promises from employers. 82 Indeed, the high costs for employees of successfully contracting around the default rules of the employment contract must often have made such rules look like immutable rules to employees. Even so, it was not the substance of these default rules themselves that constructed the employment relationship. When the default benefited the employee, after all, employers (whose costs of contracting into an alternative rule were considerably lower) often simply contracted around it, as they did when states enacted eight-hour defaults in the 1860s and 1870s. Cases such as this one show that it was not the substantive content of the default rules that determined the shape of the employment relation. With respect to the default rules, it was not the substance of the rules but often the daunting legal complexity of the law governing the interpretation and construction of contracts that did the work of promoting employer power in the workplace.	53
To be sure, employers themselves could get tripped up in the complicated rules governing what constituted enforceability in contracting out of contract defaults. Employers sometimes used language that courts construed as extending promises of employment for a term, or relied on oral promises not included within or contradictory to the terms of an integrated writing. 83 But usually technical complexity in the law of the employment contract worked to the advantage of employers. For repeat-play employers, the legal expertise required to keep track of technically complex background schemes was readily spread over the run of employees at a particular firm. And given their legal and informational positions, there was good reason for employers silently to contract into a complex and unpredictable regime of common law defaults. Doing so enabled them to keep information regarding work accident rates and business plans close to the vest. Moreover, the uncertainty generated by complex defaults was strategically advantageous for repeat-player litigants in litigation. Such complexity and unpredictability allowed firms (as Edward Purcell's exhaustive study of diversity litigation has shown) 84 to deter litigation and force favorable settlements from less powerful litigants.	54
* * *
This essay has been perforce a speculative venture. Rather than advancing a full-blown theory of the ways in which default rules in the nineteenth-century employment contract functioned, it has sought to make the more limited case that the content of particular employment law defaults did not bear a transparent relationship to the actual relations of employers and employees. Quite the opposite, the social consequences of particular defaults were most likely contingent on particular and widely varying employment contexts. Moreover, I have tried to suggest here that the presence of employee-unfriendly default rules may have been less significant than the rule complexity and unpredictability of nineteenth-century contract law adjudication. Given the differing institutional resources of employees and employers facing the prospect of litigation, employees were quite likely not to know when they were well advised to contract out of the implied common law terms. And even when employees did try to contract out, the difficulty of knowing the magic words required to do so successfully often defeated their attempts. Employers, on the other hand, were quite likely to find unpredictability and rule complexity advantageous in the event of litigation.	55
Further elaboration of these speculations, of course, will await in-depth investigations of the law in action in particular employment contexts, with renewed attention to the technical rules of the law of contracting. If the quality of the employment law scholarship of the last ten years is any measure, we can expect fine examples of such studies in the not-too-distant future.	56
John Fabian Witt is the Samuel I. Golieb Fellow in Legal History at New York University School of Law, 1999-2000, and Law Clerk to Judge Pierre N. Leval, U.S. Court of Appeals for the Second Circuit, 2000-2001. For their helpful critiques, he thanks Ian Ayres, Ariela Dubler, Barry Friedman, Bob Gordon, Beth Hillman, Anthony Kronman, Bill LaPiana, Greg Mark, Bill Nelson, Christopher Tomlins, Howard Venable, James Whitman, the members of the N.Y.U. Legal History Colloquium, and the anonymous reviewers at the Law and History Review. Notes       1. See especially Karen Orren, Belated Feudalism: Labor, the Law, and Liberal Development in the United States (New York: Cambridge University Press, 1991); Christopher L. Tomlins, Law, Labor, and Ideology in the Early American Republic (New York: Cambridge University Press, 1993); Christopher L. Tomlins, "Subordination, Authority, Law: Subjects in Labor History," International Labor and Working Class History 47 (1995): 56-90. Other important new studies of the nineteenth-century employment contract include Labor Law in America: Historical and Critical Essays, ed. Christopher L. Tomlins and Andrew J. King (Baltimore: Johns Hopkins University Press, 1992); Robert J. Steinfeld, The Invention of Free Labor: The Employment Relation in English and American Law and Culture, 1350-1870 (Chapel Hill: University of North Carolina Press, 1991); Charles W. McCurdy, "The 'Liberty of Contract' Regime in American Law," in The State and Freedom of Contract, ed. Harry N. Scheiber (Stanford: Stanford University Press, 1998), 161-97; Arthur F. McEvoy, "Freedom of Contract, Labor, and the Administrative State," in The State and Freedom of Contract, 198-235; Catherine L. Fisk, "Removing the 'Fuel of Interest' from the 'Fire of Genius': Law and the Employee-Inventor, 1830-1930," University of Chicago Law Review 65 (1998): 1127-1198. Since this essay asks questions about doctrines such as the fellow servant rule, the entire contract rule, and others, of course, it implicates a now long-standing literature on the common law of contract and tort. See, e.g., Lawrence M. Friedman, A History of American Law, 2d ed. (New York: Simon and Schuster, 1985); Morton Horwitz, The Transformation of American Law, 1780-1860 (Cambridge: Harvard University Press, 1977).
2. See Robert Hale, "Coercion and Distribution in a Supposedly Non-Coercive State," Political Science Quarterly 38 (1923): 470-94; Barbara Fried, The Progressive Assault on Laissez Faire: Robert Hale and the First Law and Economics Movement (Cambridge: Harvard University Press, 1998). Cf. James Willard Hurst, Law and Economic Growth: The Legal History of the Lumber Industry in Wisconsin, 1836-1915 (Cambridge: Harvard University Press, 1964), 286 (noting that in addition to direct allocation of resources, "official decision making also affected the allocation and use of resources at a second remove, by official determination of standards of conduct or of the range of available, legally defined and recognized procedures and instruments, within the framework of which market operations might go on").
3. The contracts literature has dedicated considerable attention to the question of default rules in recent years. See, e.g., "Symposium on Default Rules and Contractual Consent," Southern California Interdisciplinary Law Journal 3 (1993): 1-444.
4. As Ian Ayres has pointed out to me, default terms in other contexts may not require a prior contract to take effect (e.g., intestacy rules).
5. Legal scholars generally view bargaining power as an unpersuasive account of contract terms in contemporary contractual relations. Instead, the contours of such relations are said to turn on their parties' relative valuations of particular arrangements and their willingness to pay for those arrangements. At most, on this view, bargaining power explains the distribution of the surpluses generated by a transaction. See Duncan Kennedy, "Distributive and Paternalist Motives in Contract and Tort Law, with Special Reference to Compulsory Terms and Unequal Bargaining Power," Maryland Law Review 41 (1982): 563-658; Ian Ayres and Stewart Schwab, "The Employment Contract," The Kansas Journal of Law and Public Policy 8 (Spring 1999): 71-89. In the nineteenth-century employment contract, however, at least two factors made bargaining power quite important. First, workers were frequently too poor to "buy" terms, even if they might have valued them more highly than employers. Second, and more important, as a matter of practice employers (and employees, too, for that matter) rarely engaged in the "sale" of contract terms, even when it might have been rational in today's economic theory for them to have done so. Such employers operated along the lines sketched out by institutional economists early in the twentieth centry, setting basic ground rules for their employment practices and then pursuing profits within the framework of those ground rules, even when it might have been more profitable to depart from them. (The occurence of such failures of the managerial imagination, as we might style them, should hardly be surprising since we see them still today. See, e.g., John J. Donohue III, "Opting for the British Rule, Or If Posner and Shavell Can't Remember the Coase Theorem, Who Will?" Harvard Law Review 104 (1991): 1093-1119 [observing that parties to legal disputes fail to enter into deals that could often be efficient].) The frequent refusal to engage in transactions over employment terms does not, however, cut against the importance of the distinction between default rules and immutable rules. One of the central points of this essay is to show that the common law default rules of the labor relation had considerably less influence on the construction of employer ground rules than recent accounts suggest.
6. See William E. Forbath, Law and the Shaping of the American Labor Movement (Cambridge: Harvard University Press, 1991), 134-35. As Forbath explains, the effects of labor injunctions and judicial review of labor legislation were twofold. On the one hand, labor law created incentives for particular kinds of rational strategic behavior by workers. On the other hand, at a deeper level American labor law not only rewarded particular strategies, but also changed the labor movement's underlying ideological outlook.
7. For these arguments, see Amy Dru Stanley, From Bondage to Contract: Wage Labor, Marriage, and the Market in the Age of Slave Emancipation (New York: Cambridge University Press, 1998), and Steinfeld, The Invention of Free Labor, respectively.
8. Tomlins, Law, Labor, and Ideology, xv, 270 ("[I]t was law rather than nature or capital that would make the employers the masters and their employees the servants"); Orren, Belated Feudalism, 81.
9. See Hayward v. Leonard, 24 Mass. (7 Pick.) 181 (1828); see also Horwitz, Transformation of American Law, 186-88.
10. See Orren, Belated Feudalism; Tomlins, Law, Labor, and Ideology; Wythe Holt, "Recovery by the Worker Who Quits: A Comparison of Mainstream, Legal Realist, and Critical Legal Studies Approaches to a Problem of Nineteenth-Century Contract Law," Wisconsin Law Review (1986): 677-732; see also Horwitz, Transformation of American Law.