I Cannot Tell a Lie

Sarah Ngo Hamerling

As If: Idealization and Ideals, by Kwame Anthony Appiah. Harvard University Press, 240 pp., $27.95

Of all of Lewis Carroll’s fantastical worlds, one of the most enduring and impossible is a simple map. In response to a Victorian traveler’s explanation that the largest useful scale of a map is about six inches to the mile, one of his characters exclaims: “Only six inches! We very soon got to six yards to the mile. Then we tried a hundred yards to the mile. And then came the grandest idea of all! We actually made a map of the country, on the scale of a mile to the mile!” Yet the map remains folded in disuse, as farmers objected that the map would block the sunlight: “So we now use the country itself, as its own map, and I assure you it does nearly as well.” [1]

The gap Carroll charted between map and territory—between representation and reality—has since become wide enough to fit a host of enemy ideologies from diverse and far-flung disciplines. The tattered remains of Carroll’s one-to-one scale map litter the once-conquered desert, a disused fantasy of empire in Borges’ “On Exactitude in Science”; in Baudrillard’s critique of US imperialism, it is instead the territory “whose shreds are slowly rotting across the map.” [2] In the past seven years, Alan Greenspan, former Chairman of the Federal Reserve, and Michel Houellebecq, author of Submission, published books with the same title: The Map and the Territory.

Carroll’s articulation of the problem of representation—the impossibility and uselessness of a perfect representation, the necessity of an imperfect and thus strictly false map—has had a long afterlife. The image of the map and the territory occurs so pervasively, in such a wide variety of contexts, that it can fade into the problem it seeks to represent, appearing to have no content of its own. But how is it that the fable of the one-to-one map has proven so remarkably elastic, accommodating both critics of imperialism and mourners of empire, free market deregulators and post-Keynesian Marxists, abstruse postmodernists and hard-nosed economists? [3] Why is it that the map and the territory is how we pose the question of representation, and what constraints does Carroll’s formula place on its answers? What are the limits of the political economy—and the political language—that it can shelter?  

Kwame Anthony Appiah’s new book on an old problem of representation, As If: Idealization and Ideals, could easily have shared a title with the works of Greenspan and Houellebecq. Instead, he wisely steers clear of well-trodden metaphors, even as he embraces and illuminates their underlying logic. Appiah is concerned with the ubiquity of “idealization”—proceeding “as if” what we know to be strictly false were true—in diverse academic disciplines and realms of human activity. “We need many pictures of the world to do anything—to do science, to do ethics, to do politics,” he said in a recent interview. These pictures are always “going to be imperfect…it’s only useful if it leaves stuff out.” [4]

Appiah lifts his definition of idealization from the vocabulary of the philosophy of science, and his framework roughly from the work of Hans Vaihinger, a little-known Kant scholar who published The Philosophy of ‘As If’ in 1911. Appiah’s argument, at first glance, appears to be less about substance (it does not aim “to announce any startling discoveries”) than about relevance: the book offers itself as an amicable, erudite, and at times hurried 180-page advertisement for the concept of idealization as a vital topic of research beyond its narrow origins. We make simplifying assumptions in scientific models or political philosophy, Appiah explains, even though we know that gases are not colliding point masses, people are not in full compliance with norms of justice, and nation-states are not fully independent Westphalian sovereigns. We describe rational agents in economics, even though we know that real people are not optimal profit-maximizers with perfectly rational expectations. We respond emotionally to theater, even though we know that no one really dies. And, Appiah argues, we ascribe identities of race or sexual orientation to ourselves and to others, even though we hold that these categories are fictional.

Regardless of their content, idealizations may be justified—and necessary—on the pragmatic grounds that they are “useful for managing the world, including, sometimes, ourselves.” We need idealized models in so many fields not because they are true (they aren’t), but because they make the world intelligible to us, and equip us with the cognitive resources to act upon it, where a true representation would be impossible for us or unsuited to our purposes. Appiah reminds the reader again and again that the question that defines and justifies an idealization is: “For what purposes is the assumption useful?” Idealizations are the tools and preconditions of human agency.


After the Great Recession, economic idealizations have increasingly become subject to public scrutiny. From populist and leftist discourse to behavioral economics and crisis memoirs like Greenspan’s The Map and the Territory, critiques of rational choice theory as ideal—or as plain falsehood—have been in vogue. While its first and third chapters mostly summarize and reframe existing scholarship, As If presents its most original contributions in the second chapter: an extended, partially vindicating treatment of the probability theory behind rational choice, a beloved subject of Appiah’s for over thirty years.

Subjective preference and probability provide the scaffolding for modern economics. In order to model a utility-maximizing agent’s free choice under uncertainty, an economist must be able to calculate and compare her expected utility of each option. That is, we need to be able to quantify both the desirability of the possible outcomes of each option to the agent, and the likelihood of each possible outcome, as assessed by the agent.

Engaging closely with Frank Ramsey’s early twentieth century formalization of rational choice theory, Appiah convincingly argues that all attempts to render subjective probability and preference calculable rest on idealization: clearly, actual agents are not capable of these instantaneous, perfect logical calculations. In order to determine how the “content” of our beliefs—our actual preferences and probabilities—affects our actions, we must idealize away our actual cognitive capacities and constraints. But a single rational agent could never come to hold all of our inconsistent preferences and probabilities in the first place, so we must have some way of reconciling a necessary multiplicity of idealizations into the action of a single agent. In order to understand actual agents, we need a second step that, idealizing away from content, explains how the form of our beliefs—our deviations from “logical omniscience”—affects our actions.

Two successive theories, Appiah argues, are necessary to understand agency: one conditional upon the assumption that agents perfectly adhere to certain norms, and then one conditional upon the assumption that agents do not. But, Appiah implies, dual theories of normativity and error characterize more than just the market:

What we see here is the intimate connection between the description of a person or community as recognizing certain norms—a description that might be offered from the perspective of an outsider—and the understanding of those norms from the perspective of the member of the community from the point of view of the insider…. To see this is to see that an agent is not simply a thing that conforms, more or less inadequately, to the constraints of rationality that decision theory represents, but also a person who recognizes, however imperfectly, those constraints as rationally binding.

It is not a coincidence that so many of the idealizations that populate As If  “characterize the norms, conformity to which constitutes someone, for our community, as what we call ‘an agent’”: intentional minds, economic actors, sovereign nation-states, racialized and sexual persons. Idealizations are the cognitive prerequisites for our individual action on the world, but they also bear our communal definitions of who counts as an agent.

But, Appiah suggests, we identify and constitute agents not only by our idealizations—the norms to which our agents ought conform—but by the gap between idealization and reality, norms and conformance, map and territory. Appiah’s framework locates theoretical agency in error, imperfect rationality, and normative deviance.

* * *

To err, is, after all, human, but Appiah’s account of the interplay between agency, normativity, and error opens fascinating views of economic theory that, unfortunately, he leaves largely unexplored. [5] As If lacks an explicit articulation of the theoretical complement to Ramsey’s idealization of perfect rationality: Appiah argues that such a theory, which would idealize away the content of economic agents’ beliefs and preferences to explain the causal powers of their deviations from rationality, is necessary, but fails to himself supply it.

Historically, this theory was produced by Ramsey’s colleague and friend, John Maynard Keynes. Ramsey had originally formulated his theory of probability in response to Keynes’ Treatise on Probability, and Keynes’ magnum opus, The General Theory of Employment, Interest, and Money, in turn places at its textual and theoretical center a theory of how economic agents recognize but deviate from normative rationality. Of orthodox economics, Keynes writes:

At any given time facts and expectations were assumed to be given in a definite and calculable form; and risks, of which, tho admitted, not much notice was taken, were supposed to be capable of an exact actuarial computation. The calculus of probability, tho mention of it was kept in the background, was supposed to be capable of reducing uncertainty to the same calculable status as that of certainty itself. [6]

However, Keynes continues, the full assimilation of uncertainty to the structure of rational calculation is logically impossible. By uncertainty, Keynes did not mean “risky” games of roulette, life expectancy, and (to some extent) the weather—phenomena that are not certain but merely probable. Uncertainty instead refers to “the prospect of a European war… or the price of copper and the rate of interest twenty years hence, or the obsolescence of a new invention, or the position of private wealth owners in the social system in 1970.” Major political upheavals, future conditions of exchange, changes in the means of production, and outcomes of the class struggle are all incalculable, hence unknowable, hence possible—and realizable in action, for it is precisely “human decisions affecting the future, whether personal or political or economic, [that] cannot depend on strict mathematical expectation, since the basis for making such calculations does not exist.” [7]

Nevertheless, “the necessity for action and for decision” in the market compels Keynesian economic actors to act “exactly as if we had behind us a good Benthamite calculation of a series of prospective advantages and disadvantages, each multiplied by its appropriate probability, waiting to be summed.” Instead, economic actors (1) largely ignore future uncertainty, assuming that the present is a better guide to the future than they know it to be; (2) assume that the present state of opinion, expressed in prices and output, correctly sums up future prospects; and so (3) fall back on the communal or conventional judgment of the market. Expectation under uncertainty is a social relation, explicitly mediated by the normative community Appiah posits as central to the construction of agency.

Moreover, in Keynes, it is the constitutive nature of error—the impossibility of perfect rationality—to economic agency that subtends the crisis potential of capitalism: “a practical theory of future based on these principles… is subject to sudden and violent changes. The practice of calmness and immobility, of certainty and security, suddenly breaks down.” [8]

As If instructs us that while we need idealized theories to understand and act upon the world, our theories and our communities define their agents where they leave open the possibility of non-ideality, the necessary deviation from normativity. For Keynes, this is in the figure of the investor, the actor who makes choices about the allocation of capital when adherence to rational calculability is impossible. [9] For Marx, this is the proletariat, the subject of history and manifestation of agency.

The reason why we care about who gets to count as an economic agent isn’t that we have an abstract commitment to a visibility politics of the oppressed. What really matters is that economic theory has an exhortative function; the erring capitalists and workers of Keynes and Marx are not just theoretical agents, but the intended audience for the General Theory and Capital’s calls to political action. Marx, of course, ultimately addresses his writing to the proletariat. Keynes, meanwhile, directs the Concluding Notes to his General Theory to “practical men… civil servants and politicians and even agitators” who might enact his prescriptions of countercyclical deficit spending and stock market reforms (the “euthanasia of the rentier”). [10]

However, As If fails to address a reader. Appiah is “interested in the role of untruth in thinking about reality, not in the usefulness of speaking untruths.” The distinction he intends is one between cases where the idealizer knows that what they are thinking is false and cases in which they are deceived, the difference between the necessity of idealization and the expediency of deception. But all communication is not deceit: As If leaves open the urgent question of what we talk about—and whom we talk to—when we talk about ideals.

The norms that structure who counts as an agent are undeniably political—questions of personhood, disenfranchisement, and citizenship, in their broadest senses, which are expressed and negotiated in political discourse, institutional structures, material deprivation, and violent exclusion. As languages of agency, idealization and political discourse alike are languages of counterfactuals: they express what is not, but might be. “It’s only because we can understand what it would be for the world to be different from the way it is,” Appiah writes, “only because we have epistemic access to possible worlds, if you like—that we can build idealizations.”

But politics is concerned not with false antecedents but their possible consequents. The untruths we tell about actuality matter less than the possibilities—and the actions necessary to achieve them—that they illuminate in our political futures. Debates in politics, unlike in political philosophy or academic economics, are not usually fought over alternate modeling assumptions. Conversely, the invocation of modeling assumptions in political discourse serves not an opening of a conversation about economic theory, but as an implied commitment to political action.

Yet Appiah equivocates between these two senses of possibility, between what is a possible (but false) modeling assumption, and what is politically possible or attainable (but not yet realized) given our assumptions:

The questions are often complex, empirical, social scientific questions about human possibility: what changes […] is it feasible to bring about, given the way people are psychologically, given the social structures in which they are embedded? But in all these domains, in taking something false for true, we are engaging in what is, at least from one angle, our most astonishing human capacity: the ability to access ways the world is not but might have been.

The problem behind Appiah’s equivocation is not just linguistic—we only have one word for possibility—but logical: expanding the boundaries of what is possible in our model will determine what we see as possible in our political futures, while stating a political possibility will necessarily carry with it an implicit idealized model. Speaking one untruth precludes speaking the other.

* * *

What happens to the political language of possibility in As If is happening today. Appiah notes that “our unhappy proliferation of ‘alternative facts’ is not what Vaihinger”—or Appiah—“had in mind” in their thinking about useful untruths and idealizations. Nevertheless, As If’s language for describing idealization characterizes the language of the current moment: a strong notion of brute truth, a focus on identifying and adjudicating untruths. Alternative facts, fake news, post-truth politics; the words of our vigilance are drenched in truth, and the limits and aims of our political language have shifted accordingly. Our preoccupation with truth, intended as a safeguard against Trump’s linguistic rot, has led inadvertently to a different sort of decay: the hollowing out of a political language of possibility, and of the awareness of the rhetorical functions of idealization.

The pure possibility that has long characterized political rhetoric, both mainstream and radical—Yes We Can, A Future to Believe In, Another World is Possible—now sounds tone deaf. This is not about optimism. It is undeniable that material realities in the current political moment have reduced what appears as immediately possible; what’s happening to political language reflects (and does not cause) what’s happening to politics, and to the most marginalized.

But something nonetheless changes in the political language of possibility when politics becomes about hard truths and untruths. In the Language of the Third Reich, the Jewish-turned-Protestant philologist Victor Klemperer observed the corrosive drift of language around him: “The Third Reich coined only a very small number of words in its language, perhaps – indeed probably – none at all. But it changes the values of words and the frequency of their occurrence... and in the process steeps words and groups of words and sentence structures with its poison." [11] Commentators are self-righteously attentive to the perversions of language foisted on us by the Trump administration, as if truth were a buoy to cling to. But in doing so, we have swapped out the political agents of our language and our economic theories for jurors, leaving only the incredulous epistemic possibility of a fact-finder. Is it possible that the inauguration audience was the largest ever? Is it possible that the wall can be built in a year? Is it possible that Uber will provide living wages? Is it possible that tariffs will bring back manufacturing jobs?

But the point of articulating an untruth in politics is not whether we believe it, but about what its effect is when spoken—who the audience is, and what possibilities they are urged to enact. Unlike Keynes and Marx, whose theories ascribe theoretical agency to and demand political action from their audience, Trump’s untruths—to the extent that they can be constituted as a “theory”—address the working and middle classes, even as his policies are aimed to benefit the elite.

The language of political possibility is fragile, elusive in the best of times. It can sound wonkish, uncertain, or gullible. But at its most rhetorically powerful, programmatic usage slips into another modal category: necessity. We speak untruths to compel action, and to compel it now: saying another world is possible means we need another world. As an accidental cartography of the truth and untruth that occupy our political attention, As If can help us understand how to recover a language of urgency in the possibilities that are left to us. Any political possibility rests on a false representation of its subject, and Appiah is right to emphasize that shifting our normative constructions of agents in a community is what shifts the boundaries of the possible. We see this in movements for immigrants’ rights: the contestation of citizenship and community, “no borders, no walls” as possibility and as demand. We see it, too, in courtroom and workplace battles over what counts as work and who counts as a worker in an economy increasingly defined by subcontracting, the “gig economy,” and other structures both hostile to labor organizing and outside many worker protection laws.

The limits of As If are the limits of idealization. They are the limits of a political discourse in which truth evacuates possibility. Appiah asks of idealizations, “For what purposes…?” We might ask this ourselves. For what purposes have we made a map of the territory if we don’t know where we can go?  

[1] Lewis Carroll, Sylvie and Bruno Concluded (Macmillan, 1893), 169.

[2] Jean Baudrillard, “Simulacra and Simulation”, in Jean Baudrillard, Selected Writings, ed. Mark Poster (Stanford University Press, 1998), pp. 166-184.

[3] Joan Robinson, a British post-Keynesian who attempted to assimilate aspects of Marx’s work into a Keynesian framework (An Essay in Marxian Economics, 1942), is often quoted: “A model which took account of all the variegation of reality would be of no more use than a map at the scale of one to one.” Essays in the Theory of Economic Growth, London: Macmillan (1962), 33.

[4] Kwame Anthony Appiah, “Interview with Kwame Anthony Appiah – Pt. III,” Interview by Daniel Kodsi, Oxford Review of Books, http://the-orb.org/2017/09/22/interview-with-kwame-anthony-appiah-pt-iii/

[5] And, for that matter, sociopolitical structures. What, for instance, might this methodology tell us about Western constructions of gender and sexuality, if agency is coextensive with not just the acceptance of norms as binding but also with the inevitability of noncompliance? If a theory of law-breaking is constitutive of legal agents, what does this say about international law without enforcement mechanisms, or about the relationship between the racialized carceral subject and the citizen? Appiah briefly considers sexual orientation and race as idealizations in As If (and at length in his Ethics of Identity), but does not explore the connection with his account of agency and normativity as explicitly as might be hoped. Appiah writes, “Simply believing there really are or really aren’t any homosexuals is not yet to have a moral belief.” What we are to do with this perspective—what expressing, individually or collectively, that there are, strictly speaking, no races or sexual orientations makes politically possible—is left (at least in As If) underdetermined.

[6] John Maynard Keynes, “The General Theory of Employment.” The Quarterly Journal of Economics, 51, no. 2 (Feb. 1937): 213.

[7] Keynes, The General Theory of Employment, Interest, and Money London: MacMillan & Co, Ltd., 1957, 162.

[8] Keynes, “The General Theory of Employment.” The Quarterly Journal of Economics, 51, no. 2 (Feb. 1937): 215.

[9] The worker is not afforded the same status: remarking upon the “actual behaviour of labour” in striking upon decreases in nominal wages but not in the cost of living (i.e. the distribution and not the level of real wages), Keynes writes, “It is fortunate that the workers, though unconsciously, are instinctively more reasonable economists than the classical school.” Workers’ strikes are taken not to be intentional economic action as such (“unconsciously, instinctively”) but as something of an empirical quirk.

[10] The latter is explicitly intended to minimize the distortions in prices as expression of the collective judgment of agents in a state of uncertainty; that is, to bring the state of long-term expectations closer to the idealization of rational choice.

[11] Victor Klemperer, The Language of the Third Reich (Continuum, 2006), 14.