Time for Things Labor, Leisure, and the Rise of Mass Consumption

1. Parfit has an answer, but it has nothing to do with rational self-interest. Rather, he thinks that we should be concerned about the well-being of our future selves in order to follow the general moral principle that we should always be concerned about the consequences of our actions for others.

1. For a good—and surprisingly positive, given the assessment quoted earlier—overview of the history of the life cycle hypothesis, see Angus Deaton (2005).

1. The pattern reported by Deaton also casts doubt on the suggestion that people choose earnings over free time in order to leave a legacy to their heirs.

1. Carter 2006, Table Bc523–536: Enrollment in institutions of higher education, by sex, enrollment status, and type of institution: 1869–1995.

1. Interestingly, in contrast to the recessions of the 1970s and 1980s, the downturns of the early 1990s and late 2000s did not produce an equivalent spike in savings.

1. Angus Maddison (1992) observes that the pattern in the United States for savings rates not to increase as incomes go up is unique among developed capitalist economies. Keynes’s prediction about the relation between income and saving is more accurate with respect to other advanced economies.

1. On page 31 of The General Theory, Keynes (1935) states that “the marginal propensity to consume [is] weaker in a wealthy community” and that “with the growth in wealth [comes] the diminishing marginal propensity to consume.”

1. Those savings could either add to aggregate efective demand, as investment, or be hoarded. Keynes thought that ensuring the former outcome should be the primary goal of macroeconomic management in modern money economies.

1. The first was Diners card in 1949.

1. Ofer here is explicitly making a Ulysses and the Sirens–type argument, in which taking on credit works as a self-binding mechanism (Elster 1979).

1. The notable exceptions were the Depression years of the 1930s, during which consumption declined and savings increased somewhat, and, to a much lesser extent, the period of recession in the mid-1970s.

1. The whole idea that flexible production has advantages over mass production, for example, is that capital investments and capital stocks are kept lower in the former than in the latter, which facilitates greater flexibility in an unpredictable world.

292

Notes to Pages 64–72

1. Moreover, the great majority of cars were bought using credit, which would have been risky under conditions of uncertain employment (Olney 1991).

1. The value of private residential assets in 1998 dollars increased from \$100.5 billion in 1925 to \$9,183 billion in 1998. Measuring the change using a chain-type quantity index indicates a more moderate, yet still very significant increase from 17.23 to 105.01 (see the table in Appendix 1).

1. The 1913 reforms made all interest payments tax deductible. According to Prasad it was in the 1930s that promoting homeownership became an active project for the state.

1. Data collected by the National Association of Realtors indicates that the median length of tenure for sellers in the United States between 1985 and 2008 was just six years (Riggs 2016).

1. According to Eurostat data, full-time workers in France have since 2006 consistently worked more than forty hours per week despite the thirty-five-hour law. So even though the Aubry laws established a thirty-five-hour workweek, French workers chose to work more than that in order to increase their wages. As the thirty-five-hour week established a limit beyond which overtime must be paid, the obvious interpretation of the failure of full-time work hours to fall is that French workers preferred increasing their earnings (by working more overtime) to increasing their free time. In the EU, full-time work hours since 2006 have on average exceeded forty-one hours (Eurostat 2017).

1. John D. Owen (1988), in an earlier analysis suggests that higher marginal income tax rates in Europe might account for some of the diference in hours between Europe and America.

1. Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, Spain, Sweden, and the United Kingdom.

1. Angus Maddison estimates total annual manufacturing work hours per employee for 1987 in the United Kingdom, the United States, and Germany to be, respectively, 1,763, 1,909, and 1,630. For the whole economy (and this would include many more part-time employees), the numbers are 1,557 for the United Kingdom, 1,608 for the United States, and 1,620 for Germany. So in manufacturing, German workers work 15 percent fewer hours than in the United States, while across the whole economy hours per year were actually higher in Germany than in America (Maddison 1991).

1. Workers are those who in a given year have worked at all, so the term would exclude the unemployed but include the underemployed.

1. The total amount changed over time, but, somewhat oddly, the aggregate numbers in Europe and the United States tracked one another (Burda, Hammermesh, and Weil 2006).

1. The Burda study, which was based on data from time-use diaries, indicates that the amount of recorded leisure time in the sample of European countries and in the United States difered very little.

1. In survey research conducted at the turn of the twenty-first century, European workers expressed a preference for higher income at constant hours over constant income at reduced hours (Bell and Freeman 2001).

1. Keynes’s 1930 prediction was based on a projection into the future of the 30 percent reduction in work hours that had occurred over the previous fifty years in both the United

Notes to Pages 73–78

293

States and Europe. Although in Europe work hours continued to decrease through the end of the twentieth century to a greater extent than they did in the United States, the rate of decrease was substantially less than that experienced between 1870 and 1930. So, as with the United States, with increasing afluence, the rate at which hours declined in Europe decreased.

c h a pt e r 4 : a t h e o r y o f m a s s c o n s u mpt i o n a s w a g e -l a bo r c o mme nsu r a t io n

1. A growing body of work in the social and historical sciences has drawn attention to the importance of processes of measurement and commensuration in the making of modern society. Wendy Esperland and Mitchell Stevens (1998) give a useful overview of the issue from a sociological point of view.

1. The literature on this topic is a vast one, including much of the social history and historical sociology stemming from the work of E. P. Thompson, and other historians. My account here is extremely schematic and abbreviated. The point here is to give a very broad overview of the contours of the history of labor under capitalism to adequately motivate the ensuing theory. A more fully fleshed out, although still brief, summary of this historiog raphy will appear in the next chapter.

1. See E. P. Thompson (1966, 1993). One could, in the spirit of David Lewis (2002), make the argument that conventions and customs represent equilibrium solutions to coordination problems or, following Elinor Ostrom (1990), that they are solutions to collective action problems such as the tragedy of the commons. Part of what establishes consent to a given custom might then be that it represents a compromise between a fair distribution of goods and the pragmatic exigencies of a situation—the need, for example, to ensure that common goods are not depleted to the point of exhaustion. This would entail a contractual view of convention and custom. But explaining customs in such a fashion involves a good deal of projection. To interpret traditions and customs as rational solutions to problems of collective action risks falling into what E. P. Thompson (1966) famously described as the “condescension of posterity.”

1. Gareth Stedman-Jones (1984, p. 85) notes that historians of medieval law have discovered that the phrase “since time immemorial” in practice could refer to customs in existence for as few as twenty-one years. The important point, however, is that practices were regarded as legitimate to the extent that they were supported by established norms rather than because they were appraised as being fair on an objective basis.

1. Along with the idea of a fair wage, there were also a variety of ideas about what constitutes a decent or minimally acceptable wage. These would include the notion of a “living wage” and the “family wage.” But note that these concepts stipulated a bare minimum standard for what wages should aford—essentially, providing for a decent living standard (whatever that might be thought to entail). They do not set out criteria for what makes a wage a fair one, once that minimum standard had been met. Although the ideas of a living wage and family wage had great rhetorical power, they provide no guidance about what should happen when productivity increases—whether and to what degree increasing wealth should be returned to labor. The conception of fairness in economic exchange underlying the notion of the living wage is continuous with earlier notions about the right to subsistence, with roots in the preindustrial moral economy (Thompson 1991). Perhaps for this reason it works very imperfectly as a normative guide to distribution in the context of the very dynamic growth conditions, characterized by a strong trend for the productivity of labor to increase, created by capitalism.

294

Notes to Pages 79–81

1. In addition, if a set of diferent goods and services is ofered by one party, the contents of that set must be aggregated, and for that to be possible they must be made commensurable.

1. This is not to say that the objective conception of fairness is not in a very general sense itself also conventional—the proposition that fairness obtains where there is an equal exchange of commensurable substances is clearly not self-evidently true but rather amounts to a normative claim. But the convention of objectively fair exchange—according to which fairness is a function of the objective properties of the sets of things being exchanged—is neutral with regard to the social identities of exchanging parties in a way that is not the case in the example of economic exchanges between feudal lord and various dependents. In feudal exchange, for instance, the dependent is fairly treated where the lord does what is expected of him according to custom, given their respective social identities.

1. This is a basic Euclidian mathematical principle.

1. The economic consequences of that notion of fair pay have been explored by George Akerlof and Janet Yellen (Akerlof and Yellen 1990).

1. I use the term wage earners (as opposed to the more loaded word workers) to mean, quite literally, all those who exchange labor for a wage. This includes service workers and middle-class salaried workers in addition to members of the traditional working class.

1. Mandel’s account of the roots of mass consumption is that the commodification of working-class culture deprived communities of the capacity to spend free time in cultivated ways. Traditional practices and pastimes were replaced by a very passive and object-oriented style of consumption, well-suited to the needs of advanced industrial capitalism. Yet it is not clear, if indeed Mandel’s account is correct, why working-class communities should have proven so pliable. As discussed in Chapter 2, M. Aglietta (2000) gives a diferent account of how consumption norms and mass production are connected. He follows Marx in arguing that after limits were placed on the length of the working week, constraining the accumulation of absolute surplus value, businesses intensified the labor process in order to extract relative surplus value. This intensification of work then made greater demands on workers, who consequently had to consume more in order to reproduce their capacity to provide labor power. The rather reductive assumption in Aglietta’s model is that consumption is driven by the need to reproduce the capacity to labor. In efect this takes consumption to be equivalent to the actual realization of use value. However, it is dificult to make sense of the pattern sketched in the last chapter with that very limited account of consumption. Nor, indeed, does Aglietta explain why labor acquiesced in fixing the working week at around forty hours rather than fighting for further increases in productivity to be used to cut working time, at constant levels of intensity of the labor process, and at constant real wages.

Notes to Pages 81–85

295

1. There is some suggestion in Keynes that animal spirits, and the associated optimistic outlook, to some extent generate a self-fulfilling prophecy. Yet it is not clear why that should be the case.

1. Keynes’s preoccupation in The General Theory with demand shocks of the sort that led to the Great Depression perhaps distracted him from the crucial role that positive projections about latent consumer demand, made on a rational, inductive basis, play in inducing productive investment of profits.

1. The dificulty is compounded by the tendency for norms of economic action under capitalism to be specified in private terms, as rules the satisfaction of which is a matter for individuals to discern, based on introspection about things like preferences. When commensuration is understood in utilitarian terms, for example, as the disutility of work being ofset by the utility of consumption, each individual must come to his or her own judgment about what amount of consumption provides commensurate compensation. I discuss this issue in the conclusion.

1. The latter two are connected, insofar as the importance of securing the purchasing power of the wage became a matter of widespread concern. But it would be wrong to reduce supply-side changes in products and distribution to changes in the moral economy linking work and consumption.

1. The development of consumer society and the history of the institution of wage labor are related such that the two are, on the one hand, suficiently independent to have discrete identities, but nonetheless, are interdependent to the degree that change in the historical trajectory of one is highly likely to cause a change in the other and, further, that the point of contact between the two is such that the contradictions and tensions internal to one tend to get expressed in changes of form in the other. Note that the claim here is not that one line of historical development supervenes upon the other, such that there cannot be a diference in that line of development without there being also a diference in the other (Davidson 1980). Rather the two are ontologically distinct, yet systemically highly embedded in one another.

1. Although Campbell is a historicist, his ideas are in some ways similar to psychoanalytic theories according to which desire primarily seeks to sustain itself—such that desire desires desire rather than its closure, achieved through consummation. For Campbell, the point of modern consumer desire is anticipation rather than actual consumption. Yet if the object of desire is the state of anticipation, why should people ever end that state by actually acquiring what it is that they want?

1. In Aristotle’s original formulation, which Marx both developed and criticized in chapter 1 of the first volume of Capital, the exchange circuit C-M-C has the telos of acquiring goods in order to put them to use in qualitatively distinct ways. Objects serve as means to particular ends attached to kinds of activity, which are entirely incommensurable with one another. Although Marx disagreed with what he took to be Aristotle’s contention that the exchange ratio between diferent Cs is set by a combination of exigency and convention (“a makeshift for practical purposes,” Marx 1981 [1867] p.40), holding that there must be some underlying substance that determines the ratio at which diferent commodities exchange, he agreed with Aristotle’s argument that C-M-C is regulated by demand for use values that meet real wants and needs. However, the puzzling patterns identified in Chapter 2 and Chapter 3 cast some doubt on the idea that, at least under capitalism, the C-M-C circuit is regulated by real needs for actual use values, realized in actual activity (Aristotle 1981; Marx [1867] 1981; Meikle 1992, 2000; Booth 1991).

296

Notes to Pages 87–96

1. For Marx subsistence standards—understood as the minimal level of wages that induces potential workers to sell their labor power—are historically variable and so can change over time. But I would suggest that a distinction can be made between a wage level that allows for brute biological reproduction of labor power, and one that allows for a material standard of living suficient to secure a willing and quiescent labor force. For capitalism, ensuring the latter is a structural necessity. For labor, securing the former is existentially necessary. But as we move further up a hierarchy of wants and needs, consumption levels become, from the point of view of labor, increasingly negotiable, and so the question of the trade-of between real income and free time becomes more salient.

1. It might well also be that some nonwage sources of income are equally or even more likely to get spent on goods than wage earnings—windfalls such as unexpected inheritances or gambling winnings for example. In her ethnographic study of gambling, anthropologist Natasha Schull found that the unearned character of money obtained as winnings induces gamblers to gamble it away, continuing to play until their winnings are entirely depleted. Maybe the motive is to return the money whence it came, rather than withdrawing it and using it for other purposes, in order to restore a kind of moral equilibrium to the pecuniary universe (Schull 2012).

1. Indeed, this is the idea behind Zip cars, civic bike loan schemes, and similar enterprises.

1. As society becomes progressively segmented, so cultural codes become diferentiated along those segmented lines. In a complex and fragmented society, people have to negotiate code-switching between what Anne Mische and Harrison White (1998) have characterized as the discrete “netdoms” that constitute the structure of social space. One consequence of diferentiation is, therefore, increasing need for semiotic flexibility. If goods have quite local semiotic value, then being able to change them rapidly, as is the case where they are rented, enables greater ease of movement across social space.

1. The question has particular force if one conceives of households, as does Gary Becker, as analogous to firms, deploying durables as firms make use of capital equipment, in order to, as far as is possible, generate optimal levels of an output—household utility—much as capital is used to produce goods with the aim of maximizing returns to investment. Since tastes are complex and mercurial, it would seem rational for consumers not to sink too much capital into fixed means of satisfying them and, therefore, to rent goods as need and desire dictates rather than own them. If the analogy between household and firm has anything to it, why, therefore, is the market for rented consumer durables relatively small compared to the market for purchasing goods outright?

1. As noted in Chapter 2, the on average quite short period of property ownership between moves, in conjunction with high transaction costs of buying and selling homes, suggests that for most people owning is not primarily about securing good returns to an investment. Rather, owning a home seems to be, to a significant extent, an end in itself.

Notes to Pages 97–100

297

1. Indeed, housing, construed as a consumer durable, is one very prominent exception to the marginal position of the market for rented goods—although renting housing is less common in the United States, it is the norm in many parts of continental Europe.

1. Esping-Andersen uses the term decomodification rather than commodification because all of the countries he analyzes are capitalist, so the default state is for labor to be commodified. The action of diferent welfare state regimes to difering degrees decommodifies this antecedent state of afairs.

1. The liberal group is Australia, the United States, the United Kingdom, New Zealand, Canada, and Ireland; the corporatist group is Italy, Japan, France, Germany, Finland, and Switzerland; and the social democratic group is Austria, Belgium, the Netherlands, Denmark, Norway, and Sweden.

1. The OECD countries in the corporatist group consist of France, Germany, Italy, Japan, and Switzerland; the OECD social democrat group, of Austria, Belgium, Netherlands, and Sweden; and the OECD liberal group, of the United States, the United Kingdom, Australia, and Canada.

1. At the same time the countries of continental Europe have shorter working weeks, and renting homes, as opposed to homeownership is more common than in the liberal states.

1. For De Vries the direction of causality between wage labor and consumer demand is the reverse of the argument of this book—the desire to consume prompted an “industrious revolution,” the result of which was greater participation in labor markets.

1. And indeed, there is at least a loose correlation between early periods of capitalism, in which the putting out system predominated and what many historians have referred to as a “traditional,” or “customary” attitude to work and time existed, in which the inclination to work was either relatively insensitive to changes in pay rates or else declined as hourly pay rates increased (Thompson 1991; Hobsbawm 1964; Gutman 1973).

1. According to this theory, humans are endowed with two diferent thought systems, system 1 and system 2, which developed at diferent points in our evolutionary history. System 1 is fast and makes use of heuristics, while system 2 is slow and involves reasoning (Kahneman 2011).

1. As philosopher John Rawls (1999, p.  293), expressing the intuition underlying the normative objection to hyperbolic discounting from the point of view of rationality, puts it: “Rationality requires an impartial concern for all parts of our life. The mere diference of location in time, of some things being earlier or later, is not a rational ground for having more or less regard for it.”

1. On the other hand, perhaps the prospective ratio of what Frank Knight (1921) calls risk and uncertainty changes over future time in a quite nonlinear fashion. In Knight’s definition, risk involves measurable probabilities, like the chance of getting a six on a role of a die. Uncertainty, on the other hand, refers to situations in which we have no measurable probabilities. Perhaps in the short term, risk predominates, but, at a certain point in the future, an epistemic horizon is reached, making it impossible to assign probabilities on a rational basis. At that point the world switches quite abruptly from being risky to being uncertain. If any two given days more than, say, a year in the future will be equally dominated by uncertainty, then there would not be a good reason to distinguish between them. They would be equally “unknown unknowns.”

298

Notes to Pages 100–108

1. More on this point follows, in the discussion of possible objections to the theory.

1. Note that the finding is not that the young do not borrow but rather that they do not borrow enough to fund discretionary consumption at levels that would smooth consumption over the life course.

1. A sense of urgency to turn money into things, in order to make it commensurable with labor, perhaps explains the feeling, and consequent pattern of behavior, described in the adage that “money burns a hole in one’s pocket” (the earliest origins of this adage, according to the Oxford English Dictionary, can be found in the eighteenth century “My tyme lyes heavy on my hands, and my money burns in my pocket” G. Farquhar [1702], Inconstant v.iii 77 [OED]). What burns is the memory of a deficit, constituted by free activity sold as work. Spending money, then, is an attempt to fill this deficit and turn what might well have been experienced as an exploitative relationship into one retrospectively becalmed by the projection of an aura of equal exchange. The saying “work hard, play hard” is also suggestive with respect to the commensuration theory, as it evokes an equivalence between work and leisure—that the two need to be in some sense of equal intensity for them to be in a balanced relationship.

1. It is perhaps not surprising that the struggle over hours took place mainly in an era before wage labor had become fully legitimate, a quasi-naturalized fact about social life.

1. In acquiring things, people generally do not acquire more free time. But there are two possible exceptions to this. The first is certain labor-saving devices. However, as we saw in the last chapter, the empirical evidence suggests that the deployment of such devices has not freed up much more discretionary time (Cowan 1983; Robinson and Godbey 1999). The second exception is artifacts that increase life span. To be sure, warm clothes and adequate heat and nutrition can increase the life spans of the poor. But once afluent conditions prevail, it is not clear that further increments of consumption lead to further increases in life span. What of consumption directed to health care? An argument could certainly be made that spending on health care increases life span. Life-prolonging devices such as pacemakers clearly add time to lives, and it might well be the case that some part of the preference for higher wages over less work time is explained by increasing health care costs, at least in the United States. But it is doubtful that the general pattern described in the last chapter can be explained away in that way. The postwar pattern for real incomes to increase and for decreases in work time to slow to a crawl was also experienced in other advanced industrial capitalist nations. For example, it is also evident in the United Kingdom, which, from 1945 onward, has had a national health care system that makes access to health care independent of income. Furthermore, a very significant determinant of life expectancy in advanced industrial countries is lifestyle. Stress, lack of sleep, not having time to exercise, and eating badly all have a pronounced negative impact on life expectancy, and all are associated working conditions under capitalism. Given those correlations, working less at constant levels of income, if it leads to decreased levels of stress, might well yield a health and longevity dividend that more than ofsets the added longevity yielded by more health care at the expense of a longer work week.

1. Perhaps this hypothetical plenitude of use value underlies the mistake of conflating increasing wealth with increasing well-being. The accumulation of material wealth entails (the service economy notwithstanding) the accumulation of commodity objects. But wealth is plausibly construable as a measure of well-being only if it can reasonably be expected that the useful powers subsisting within it will be deployed to improve lives. Otherwise, wealth remains inert, unused, and therefore useless. To see wealth as a gauge of well-being is to forget this basic point. Amartya Sen (1988) has repeatedly pointed out the perversity of this way of thinking, and the awful consequences it has for human welfare— yet this raises the question of why such odd habits of thought should have become so deeply entrenched.

Notes to Pages 109–115

299

1. The shift to services was to some degree facilitated by globalization, as industrial production moved elsewhere in the world. Although the proportion of workers in manufacturing might have gone down in the West, it increased dramatically elsewhere in the world, as people moved from agriculture to industry. Likewise, although the United States, for example, produces relatively fewer material commodities than used to be the case, the rate at which consumer commodities are accumulated in the United States has remained as high as ever. The only diference is that many of those commodities are now imported.

1. The general tendency to hoard then might be explained by aversion to loss of goods that are inscribed with the sacrifice involved in the labor exchanged for the money used to buy them. David Lord Smail (2014) has suggested that hoarding has both a neurological and a cultural-historical basis and that the study of the epigenetics of the behavior might be a fruitful way to bring the two together.

c h a pt e r 5 : e c o n o mi c f a i r n e s s a n d t h e w a g e l a bo r bac kg r o u n d

1. Some psychologists, especially those inclined toward evolutionary explanations, hold fairness in a very general sense to be a transhistorical category, perhaps even extending beyond the boundaries of the human species (Humphries 2006). These analysts draw on experimental evidence that indicates that fairness plays a major role in social behaviors to support the claim that “it is in the nature of human beings to value objective fairness” (Mack 2006). The suggestion is that a sense of fairness is part human nature, a component of a “moral instinct” that is inscribed in our genes. Be this as it may, the diversity of cultural conceptions of fairness evident across time and space suggests that if there is an invariant substrate to the instinct to be fair, it is rudimentary and underdetermining. As evolutionary game theorist Ken Binmore (1994) has observed, even if it is the case that people have a “fairness gene,” how it gets expressed is clearly mediated by, and therefore contingent on, historical and cultural context.

1. This is not to say that the idea of objective equal exchange itself was new with modernity. Aristotle (1982), in book five of the Nichomachean Ethics, considers the question of fair exchange at length. But Aristotle’s philosophical inquiries were not the basis of an ideology governing economic activity in the ancient Greek world.

1. For example, in the feudal system lords would grant land to vassals in exchange for military service, while peasants gave labor services in exchange for protection and the right to work the land (Bloch 1961).

300

Notes to Pages 116–124

1. As I argued in the last chapter, one can make a distinction between the internal legitimacy of an institution such as feudalism—which is a matter of the degree to which the institution in practice lives up to its normative ideals—and the legitimacy of a normative system itself, which can be challenged on the external grounds that it violates other norms and customs, or is in tension with changing expectations.

1. Here I’m invoking the distinction made by Karl Polanyi (1968) in his analysis of market society.

1. Wierzbicka argues that there is no word in other European languages exactly equivalent to the English term fair. Insofar as it was in an English-speaking country that industrial capitalism first developed, her findings are consistent with the argument that the modern concept of fairness had something to do with the emergence of capitalism. Moreover, Wierzbicka admits that this concept is nowadays widespread in modern capitalist societies. It should be noted that other linguists have disputed Wierzbicka’s suggestion that the modern, ethical sense of the English word fair cannot be found in other languages.

1. As Wierzbicka comments (2006, p. 152): “values like ‘equity’ and ‘equal distribution of goods’ have become so important for many people that they have come to see them as the essence of ‘fairness,’ and when they try to define fairness, many people do so along the lines of ‘equity.’”

1. The matter is complicated, however, by the degree to which a given exchange is seen as taking place under coercive conditions. If a given exchange were unequal, yet both parties agreed to its terms under completely noncoercive conditions, would it be regarded as unfair? The question raises complicated issues. Objectively unequal exchanges that are knowingly and noncoercively entered into might not be categorized as a narrowly economic exchange—the kind that occurs in markets—and instead fall into something like charity or gifting. On the other hand, some theorists subsume gifting under an expanded concept of formal economic exchange (Akerlof 1982).

1. The list of historians whose research has shown the degree to which legitimacy was not accorded to the institution of wage labor in the early years of capitalism is extremely long. It would include E.  P. Thompson, Eric Hobsbawm, and other luminaries of British social history, as well as pioneers of American social and labor history such as David Montgomery, Sean Wilentz, David Roediger, Eric and Philip Foner, and Herbert Gutman.

1. It was only in the last quarter of the nineteenth century that wage earners in America finally came to outnumber the self-employed.

1. In part, this was because employers often specified very stringent terms in their labor contracts. For example, Stanley notes that the so-called ironclad contracts drawn up by employers, which forbade collective bargaining, were a cause of much ire among workers. The rise of the written labor contract laid bare class divisions and conflicts of interest, became a focal point for industrial action, and so propelled the “labor question” into national consciousness.

1. The phrase wage slavery was not unique to America—it was also used by the Chartists in Britain and by Friedrich Engels, in his journalism.

1. Indeed, the Adkins ruling was in efect overturned with the passage of the Fair Labor Standards Act of 1938.

Notes to Pages 124–130

301