The Dismal Science Project
|
The Papers
|
The Images
|
||||||
|
Post-Ricardian
British Economics, 1830-1870
|
|
||||||
|
Analytical
Egalitarianism,
“The difference between the most dissimilar characters, between a philosopher and a common street porter, for example, seems to arise not so much from nature as from habit, custom, and education. When they came into the world, and for the first six or eight years of their existence, they were, perhaps, very much alike, and neither their parents nor playfellows could perceive any remarkable difference. About that age, or soon after, they come to be employed in very different occupations. The difference of talents comes then to be taken notice of, and widens by degrees, till at last the vanity of the philosopher is willing to acknowledge scarce any resemblance.” Smith (1976, 28-9). Analytical Egalitarianism When Adam Smith attacked the doctrine of innate differences of people, he singled out the “vanity of the philosopher” for his belief that he was somehow of a different species than a common porter.1 Ordinary people, says Socrates in Plato’s Republic, cannot be philosophers because they remain content with many surface phenomena whereas the philosopher inquires after the one explanation.2 Since Plato’s time, surface phenomena have acquired a label: anecdotal evidence. The difference in knowledge between philosophers and ordinary people is key to a class of criticism of democracy: ordinary people bring nothing to the political process other than their many anecdotes, evidently inferior to the expert’s model. Similar arguments have been made to immunize “science” from criticism by ordinary people: for, if ordinary people cannot abstract from surface phenomena, they have no basis (except anecdotes) upon which to criticize the scientist.3 Responding to this criticism in the 3 political theory context, workers in rational expectations politics suppose that while experts have a deeper understanding of the process, ordinary people can by heuristics act as if they understood the model (Wittman 1995). We regard the Platonic principle as analytical hierarchicalism. In terms of knowledge, the expert dominates ordinary people. The valuable contribution of rational expectations is to show that in interesting and important contexts the expert only weakly dominates ordinary people. But we propose to go beyond the rational expectations point of view to demonstrate that information aggregation of anecdotal evidence can improve upon the expert’s model in the case where the model is not exactly correct. Thus, neither the expert nor the ordinary people dominate the other. In the language of statistical decision theory, both the expert’s model and the information aggregation process we sketch below are “admissible” estimators. There are two pieces to the argument. The technical statistical part below is a demonstration that the median of anecdotal evidence is an admissible estimator. (Lehmann 1986, p. 16) What this proves is that the expert’s model does not dominate the information aggregation possibility of ordinary people in a democratic political context where the policy is made by the median voter. Thus, we argue for analytical egalitarianism by providing a counter-example to analytical hierarchicalism. This is the first piece of the argument. Elsewhere we consider why it might be in the interest of the theorist to propose a model which is not exactly correct. This we regard as a sustained argument against analytical hierarchicalism. We argue that the theorist is motivated by the same sort of desires which 4The gendering is not random. Episodes will be described below in which the male theorists attempted to remove the participation of female onlookers. 4 motivate ordinary people (Levy 1988, Feigenbaum-Levy 1995, Levy 2001, Peart-Levy 2002). If the expert is not constrained to seek out a median anecdote or some other admissable estimate, but is instead allowed to choose any anecdote, then – as our cannibalism example below demonstrates – we find some extraordinarily awful “expert” arguments being made. We propose to develop some machinery to make precise the difference between expert’s model and ordinary people’s anecdotes in the context of democratic politics. This machinery is then employed to cast light on two historical episodes: i) the exchange between Thomas Carlyle and J. S. Mill; and ii) the allegation of cannibalism in the events in Jamaica in November 1865 known as the Gov. Eyre controversy (Levy 2001, Levy-Peart 2001). In the first case, we find explicit recognition by the political economists of our technical egalitarian argument for the median anecdote. In the second, we find an example of the worst imaginable outcome of a failure of analytical egalitarianism – when an “outlier” is presented as “truth” about the negroes in Jamaica. The signature of the analytical hierarchicalism with which we are concerned is the expert’s use of his model to persuade an ordinary person to disregard the evidence of her senses.4 By this we do not mean that the expert explains to the ordinary person that there is other sense evidence which outweighs hers, but rather what she sees does not exist because it is not in the expert’s model.
|
|
||||||
|
Denying Human Homogeneity:
“Let us bear in mind the words of Galton written almost in the last years of his life, words not of despair, but of wise caution: ‘When the desired fullness of information shall have been acquired, then and not till then, will be the fit moment to proclaim a “Jehad” or Holy War against customs and prejudices that impair the physical and moral qualities of our race.’” (Pearson and Elderton 1925, p. 4). Introduction The question we propose to address is how peace came to the conflict between economics and racism. How did economics move from the Classical period characterized by the hardest possible doctrine of initial human homogeneity – all the observed differences among people arise from incentives, luck and history1 – to become comfortable with accounts of human behavior which alleged foundational differences among and within races of people? (Darity 1995) In this paper, we shall argue that early British eugenics thinkers racialized economics in the post-Classical period. |
|
||||||
|
Sciences Gay
& Dismal:
Stories and Models The great debates over hierarchy in Victorian Britain occurred, just as Thomas Carlyle (1849, 1850) said, between supporters of the “dismal science” of economics – in conjunction with the Christian Evangelicals of Exeter Hall1 – and the practitioners of the “gay science” of poetry. In particular, when the Eyre Controversy came out of Jamaica, the British intellectual community separated along disciplinary lines – economists, biologists, and Evangelicals sided with the Jamaicans; while poets and artists defended Governor Eyre (Semmel 1962, Holt 1995, Levy 2001, Levy-Peart 2001). In the judgment of many historians, the poets wrongly triumphed. Why has art been so persuasive in competition with science?2 The answer, we shall assume, is that ordinary people understand stories better than they do models. But if so, how was classical economics ever able to compete effectively. We shall claim that classical economics had story-like (inductive) properties. Two consequences follow from the inductive procedure which presupposed human homogeneity. First, it made classical economics the natural ally of Biblical story-tellers who also assumed human homogeneity. We have discussed this coalition elsewhere (Levy 2001, Levy-Peart 2001 & 2002). Here we consider how classical economics became the natural target of later modelers who presupposed a racial hierarchy. We distinguish in what follows between policy, questions such as “how the Jamaicans ought to be treated,” and metapolicy, questions such as “how we solve the problem of ‘how the Jamaicans ought to be treated.’” We provide a framework in which various sorts of metapolicies might be distinguished, and we propose a model which might predict which sort of metapolicies will be the rational choice of a policy advocate.. |
|
||||||
|
Common
Themes in Ruskin and Galton:
Introduction The title announces a challenge. Other than the fact that they spent much of their life on earth in the same part of the world, along with a few million others residents of England, just what do John Ruskin and Francis Galton have in common? A search in Karl Pearson’s magisterial 1912-1930 Life, Labours and Letters of Francis Galton turns up precisely one reference to Ruskin. Here it is from a letter of 1873 I have often thought of procuring a really artistically made and coloured globe and once had much correspondence about it. Ruskin wrote a very good letter (1930, p. 461) In fact, as we shall endeavor to demonstrate, Ruskin and Galton had much in common. What concerns us flows from their belief in the natural differences among humans. As the cofoundation of the “science” of eugenics, Galton’s assent to a hierarchical view of the human is well-known. Ruskin also accepted such a position, although this is perhaps less well-known (Levy 2001, Levy-Peart 2001/2002). Here is a characteristic passage from his attack on J S Mill’s economics: .. if there be any one point insisted on throughout my works more frequently than another, that one point is the impossibility of Equality. My continual aim has been to show the eternal superiority of some men to others, sometimes even of man to all others; and to show also the advisability of appointing such persons or person to guide, to lead, or on occasion even to compel and subdue, their inferiors according to their own better knowledge and wiser will. My principles of Political Economy were all involved in a single phrase spoken three years ago at Manchester: “Soldiers of the Ploughshare as well as Soldiers of the Sword”: and they were all summed in a single sentence in the last volume of Modern Painters–“Government and co-operation are in all things the Laws of Life; Anarchy and competition the Laws of Death. (Ruskin 1905, 17:74-5) The immediate question which a hierarchicalist worldview raises is how to distinguish the better from the worse? Ex post it is easy enough to tell a plausible story about winners and losers, but if a theorist wants to use this insight in social policy, s/he needs to make this distinction operational ex ante. One obvious way by which people are distinguished is that they look different. Since anthropologists in the mid-nineteenth century focused on physical differences – and the purported inferiority of race that was associated with some such differences (Peart-Levy 2003)– it is, perhaps, not surprising that physical traits would figure prominently in the analysis of mental traits by Ruskin or Galton. |
|
||||||
| Those with an interest in literary studies might wish to read this paper on market and cultural studies. | |||||||
|
A
Series in
For every policy debate, there is is a “higher” order debate, a disagreement at the level of what, for lack of a better term, we call metapolicy. One side automatically abstracts; one side automatically particularizes. The question is what conclusion can we draw when the differing approaches to policy point in opposite directions? In the 19th conflict about which we were talking at the Cato meeting, it isn’t that “gay scientists” – Carlyle’s poets – had a model with which to combat the economists’ model of abstract economic man who remains the same as the world around changes. No, the poets told stories and wrote novels. I gave particular attention to Carlyle’s “Negro question,” a story of an astounding lecture at Exeter Hall. In the work which Peart and I have been doing, we find Charles Kingsley’s Water-Babies singularly interesting. In it Carlyle’s gospel of humanization through labor and hierarchy, taking into account developments in evolutionary biology, is retold in a form suitable for children. Unlike the “Negro Question,” Kingsley’s story has never been out of print since its first printing in 1862. In Kingsley’s telling those people who were allowed to DoAsYouLiked devolved into apes and were exterminated by their betters. And just this week, a slick new version of H. G. Wells’ Time Machine opened at the movies with its account of racial devolution and genocide as a policy option. It is perhaps because Dismal had to deal with narration on a serious level that I was asked by Chris Shea in the interview for the Chronicle review “When did you leave economics?” any systematic study of these series of exchanges must attempt to bring model and narration into some common ground. In the 19th century debates over slavery, British narrators and modelers spent a good deal of effort arguing with one another so some of my work was already done for me. Modelers reviewed novels. Richard Whately’s response to Kingsley’s and Dickens’ attack on markets occurs in his review of Uncle Tom’s Cabin in which Harriet Beecher Stowe gives voice to the anti-market opinions through her character of a benevolent slave-owner! Scholars in cultural studies, as students of literature and other forms of narration, assume that metapolicy takes the form of narration. Economists not surprisingly assume that metapolicy takes the form of model building and testing. Consequently, I believe that participants spend a good deal of effort but manage to talk past one another. I would like to propose a series – Market & Cultural Studies– which would provide an occasion for adherents of differing metapolicies – economic models and literary narration – to speak to each other and to the scholarly community as a whole. I would like to have Sandra Peart as co-editor. I think we’d bring an agreeable diversity, in more dimensions than I can easily count, to the project. If the series were co-edited then we’d have to agree. This would mean something to readers. When Steve Medema discussed a paper of ours in Atlanta he said that Sandy keeps me under control. I have been informed that this is correct. |
|
||||||
|
From Median
Well-Being to Mean Well-Being:
2 The principle of the greatest happiness may have gained its popularity, but it lost its meaning, by the addition ‘of the greatest number.’” (1881, p. 118) We would prefer to phrase Edgeworth’s point as the claim that the theory of utilitarianism has a multiplicity of models.2 One utilitarian model proposes the mean income be maximized; another which proposes that the median be maximized. As illustration of a median utilitarianism, here is perhaps Adam Smith’s most straightforward statement of this majoritarian position when he asks about the evaluation of growth: Is this improvement in the circumstances of the lower ranks of the people to be regarded as an advantage or as an inconveniency to the society? The answer seems at first sight abundantly plain. Servants, labourers and workmen of different kinds, make up the far greater part of every great political society. But what improves the circumstances of the greater part can never be regarded as an inconveniency to the whole. No society can surely be flourishing and happy, of which the far greater part of the members are poor and miserable. It is but equity, besides, that they who feed, cloath and lodge the whole body of the people, should have such a share of the produce of their own labour as to be themselves tolerably well fed, cloathed and lodged. 3 What motivates our research is that modern economists, without reflection, use the mean for utilitarian welfare evaluation. For empirical work per capita income is the standard for international comparisons; indeed, “economic growth” is defined using average income. For theoretical consumer’s surplus analysis which establishes the relative efficiency of industrial organizations, we total the gains of producers and consumers. With a fixed population of size N, the mean is simply the total divided by N.4 How did we go from one to the other without a discussion? Interest in the median seems to be restricted to workers in public choice for whom it has been long apparent (Galton 1907a&b, Black, 1953; Downs 1957; Plott 1967) that the median is the natural centering principle for a majority-rule democracy.5 How does one move from a mean-based norm to a median-based positive theory without the mirthinducing assumption of symmetry? Historically there are many “problems” associated with the idea of a majority rule democracy which are immediate implications of the median. 6Indeed, one of the puzzles with which associates of Gordon Tullock have long been pestered is just how one “justifies” democracy. 4 Even in a single dimensional context there is no attention paid to intensity of preference since the median counts but does not add. In a multiple dimension context there is no ability to trade which different intensities of preference would motivate (Buchanan- Tullock 1962, Plott 1967). It is easy to document that the well-nigh universal use of mean income in neoclassical analysis is intimately connected with the intensity-blindness of the median. Maximizing the mean allows for a policy of compensation – the winners can redistribute their gains to make the losers no worse off – whereas maximizing the median does not allow such compensation.6 Nonetheless, the later classics, J S Mill in particular, considered compensation as a necessary condition for reform. We propose that the classics employed a richer principle which justifies compensation in which the payments need not be restricted to the material gain from the reform. This principle depends upon the fact that in classical political economy the subjects of the theory of political economy are also theorists. One postulate in the theory of ordinary people, in some reflective equilibrium, is that others have a standing in exchange comparable to ours. Adam Smith expressed this most memorably when he claimed that people trade but dogs do not because we have and they lack a concept of “fair” which for him depends upon the distinction between “mine” and “yours.”7 But what if there is only “mine” because you are only an instrument of my will? Does this state of affairs pass the reciprocity norm which the classics presupposed? (Levy 2001, Levy-Peart 2001/2002)? |
|||||||
|
Justice
Toward Another Rational Species: David M. Levy
|
|||||||
|
Econometrics
and the
Two Orders The anomaly does not disturb the placidity of textbook accounts, but the remorseless logic of self-seeking economic calculation stops at the edge of economic theorizing. The economic theorists are supposed to have motives other than those which characterize the ones we study. We who study self-seeking behavior are supposed to be truth-seekers. The term “philosopher” is Greek for “lover of wisdom.” As truth is the paradigmatic public good, there is no immediate, direct link to our self-seeking behavior. What might be the alternative to the truth-seeking assumption? We propose that economic theorizing can be most usefully modeled as an act of exchange and the theorists have the same self-interested motives as those we theorize about. We are who we study. And when we forget this we find ourselves in difficulties.1 Our argument below concerns econometric theory and practice. Econometrics is the cross-road from pure theory to that entity which is sometimes called the real world. The bridge between the platonic world of pure mathematics and the real world is, we shall argue, something described in choice-theoretic terms. The truth-seeking assumption in econometrics seems to us to explain the otherwise curious lack of a code of econometric ethics. There is, after all, a long tradition of codes of ethics in mathematical statistics, ASA 2000 being only the latest from the American Statistical Association.2 Codes of ethics can be viewed as attempts to modify the behavior of those for whom there is a difference between their interests and the interests of others.3 Recognition of the prisoner’s dilemma is the mark of the human.4 The point of the paper is that we are subject to the prisoner’s dilemma whether we recognize this or not. The truth-seeking assumption obviates the need for a code of ethics because truth is a pure public good. There can be no difference between what is in our interest and the public interest. We shall argue that this self-serving belief has drastic consequences. When we believe that everyone else is a truth seeker, and thus the literature will be free from gross, systematic error, we can free-ride on their efforts. This free riding destroys the conditions necessary to establish freedom from systematic error. Econometrics has long been seen as the most narrowly scientific aspect of economics. As such it would seem to be subsumed by what may be the controlling “linear” image of the scientific process.5 To mangle “Newton’s” metaphor, we see better than our giant predecessors because we stand on their shoulders.6 In the image the only goal visible is that of insight. What is desired is clarity into how the world works. The fact that we are able to see farther than others is purely an accident of history: we come after them. Reverse the order and they would see farther than us. Consider two scientists, G and D. The truth which each of them can recover individually is T(G) and T(D). Add their efforts together, then the scientist D who can use the efforts of G, thus D|G, can find more truth. In particular T(D|G)=T(D)+T(G). If truth is all that matters to both D and G then we have “solved” the problem of the public provision of public goods by assuming that our individuals wish, and only wish, to produce the public good of truth. What if scientific progress is not a linear affair in which the past is nested in the present, but rather zig-zags, improving in the main but with some local losses?7 Schumpeter’s unrivaled mastery of the history of economics leads him to the zigzag hypothesis in which one cannot evaluate the present without knowledge of the past.8 Although Schumpeter believes the history of science “teaches us much about the ways of the human mind” (1954, p. 5), he does not give the microfoundations of the zigzag process. Indeed, he asserts that the motivation of scientists is totally irrelevant to science. It makes no difference whether the scientist is a truth seeker or a hired expert; science is objective.9 But if the problem is to explain the systematic error, the zigs and the zags, then perhaps the motivation of scientists are important. What is the alternative to the truth-seeking assumption? We propose to remember that we are economists and explain theorizing as a trade. Thus, we propose to bring the theorist within the katallactic circle of economics.10 We use “circle” in a non-metaphorical sense. Over one hundred years ago when Fleeming Jenkin explained the logic of exchange, he drew a picture of the circular order which results from trade. The nonlinearity results from each individual pursuing her own interests. This image has been neglected so we reproduce it.11
|
|||||||
|
‘Not
an Average Human Being’: How Economics Sandra J. Peart Introduction Our earlier contribution to this volume showed how racial theorizing was used to attack the anti-slavery coalition of evangelicals and economists in mid-nineteenth century Britain. Classical economists favored race-neutral accounts of human nature, and they presumed that agents are equally competent to make economic decisions. Their opponents such as Carlyle and Ruskin presupposed racial hierarchy and argued that some people are incapable of making sensible economic or political decisions. They concluded that systematically poor optimizers will victimized in either market or political transactions. In this chapter, we shall show how the attacks on the doctrine of human homogeneity succeeded: how, late in the century, economists came to embrace accounts of racial heterogeneity entailing different capacities for optimization.1 We attribute the demise of the classical tradition largely to the ill-understood influence of anthropologists and eugenicists,2 and to a popular culture that served to disseminate racial theories visually and in print. Specifically, W. R. Greg, James Hunt, and Francis Galton, all attacked the analytical postulate of homogeneity that characterized classical economics from Adam Smith3 through John Stuart Mill. Greg co-founded the eugenics movement with Galton and he persistently attacked classical political economy for its assumption that the Irishman is an “average human being,” rather than an “idiomatic” and an “idiosyncratic” man, prone to “idleness,” “ignorance,” “jollity” and “drink” (quoted in full, below). By 1870, two theories of race co-existed in the scientific community and the popular press. The more devastating view of the owner of the Anthropological Review, James Hunt, held that there were races whose physical development arrested prematurely, dead races incapable of elevation. The second, which we we call “parametric racism,” held that the inferior race differed from the superior (Anglo-Saxons) along some parameter(s). As both sorts of racial theories entered into economics in the decades that followed, the focus moved from physical differences stressed by the anthropologists – the shape or size of the skull – to differences in economic competence. Economists argued, for instance, about whether the Irish or blacks in America were competent enough to make the labour supply choice, or to save for their old age. We shall demonstrate how pervasively these racial accounts entered into economic thinking well into the twentieth century, in economists’ characterization of the family size choice, intertemporal decision making, and the consumption of “luxuries” and intoxicants. The influence of eugenicists on economics extended to policy. As economists came to accept racial accounts of economic behavior, they allowed that some among us are “unfit,” parasites who live off of the rest of society. They endorsed an elaborate “remaking” program for inferior decision-makers, and for many economists the re-making was also to be biological. A major theme below shall be how such policies were designed to reduce the level of what they called “parasitism” in society. While eugenics is now commonly understood to have been influential but mistaken policy, the tension between economists who presume that agents are equally able to optimize, and those who wish to improve the economic competence of various groups, has never been fully resolved. Racial accounts won the day well into the twentieth century, but near the middle of the century the classical tradition of homogeneity was revived at Chicago. Not surprisingly, given the racial characterization focused on intertemporal decision-making, time preference was central in the Chicago revival. In his 1931 review of Irving Fisher’s Theory of Interest, Frank Knight voiced his skepticism about the common link supposed in economists’ accounts between time preference and race. Knight, and after him George Stigler and Gary Becker, questioned myoptic accounts of intertemporal decision making. As the Chicago school revived the classical doctrine of homogeneity it also (and by no coincidence) revived the presumption of competence even in political activity.
|
