I recently started Albert O. Hirschman’s “Rhetoric of Reaction” (a classic for a reason; beautifully written, remarkably even-handed, schematically indispensable; its faults, though real, can be reserved for another conversation.) Hirschman famously categorizes rightist (more generally, anti-change) arguments into “perversity,” “futility,” and “jeopardy” theses – arguments that the effects of a proposed change will be (respectively) contrary to its original intentions, entirely nil, or serving to eliminate some previous and more precious accomplishment. It implicitly poses an obvious and complementary question: can we propose a similar schematic for left-wing arguments? Sure! Trivially, even.[1]
As the opposing team to Perversity, Futility, and Jeopardy, then, let’s look at what I’ll call Feedback, Induction, and Expertise. Just as the first three are employed by the right across a wide variety of domains, so the second are employed by the left. Just as the first can be appropriated occasionally by the left as well, the second can find occasion to be used by the right. And just as the first three exist in a certain amount of tension and even contradiction with each other, so does the second trinity.
Like Hirschman, I am on the left, meaning that unlike him, I am attempting to explicate arguments that I broadly support (or have an incentive to.) However, like him, my at least explicit goal is neither to praise nor bury these “theses” or classes of arguments. Like perversity, futility, and jeopardy, each of feedback, induction, and expertise admits of accurate as well as inaccurate uses.
I. The Feedback Thesis
The feedback thesis is that social (dis)advantages tend to compound and reinforce themselves over time. Racial prejudice against blacks as stupid stymies their achievement in higher education, and these weaker outcomes as seen as evidence that they are intrinsically stupid. Poor health impedes one’s ability to work productively, which impedes one’s ability to earn income and thus purchase healthcare. Membership among the rich grants you political power, which you can use to keep you and your buddies rich. Examples of this could be repeated ad nauseam – indeed, it’s my claim here that, if you’ve spent much time reading anyone to the left of Skeletor, you could repeat them ad nauseam yourself – but perhaps the purest and most general form comes from Pierre Bourdieu:
Depending on the field in which it functions, and at the cost of the more or less expensive transformations which are the precondition for its efficacy in the field in question, capital can present itself in three fundamental guises: as economic capital, which is immediately and directly convertible into money and may be institutionalized in the forms of property rights; as cultural capital, which is convertible, on certain conditions, into economic capital and may be institutionalized in the forms of educational qualifications; and as social capital, made up of social obligations (‘connections’), which is convertible, in certain conditions, into economic capital and may be institutionalized in the forms of a title of nobility.[2]
Power comes in many different forms, says Bourdieu, but these can each be used to get the others. And feedback need not cross between domains either. Thomas Piketty is hardly the first to note the consequences of the fact that (literal) capital (literally) compounds upon itself.
The feedback thesis has three important components or implications – strategic, ontological, and epistemological. The strategic implication is that reality is, to a great extent, highly path dependent. Path dependency, after all, is a general feature of systems where changes are self-reinforcing, just as convergence is a general feature of systems that are self-correcting. In this “strategic” sense, feedback arguments can be understood as a general counter-argument to “futility,” as they fundamentally model a world where outside intervention can permanently upgrade things from lower to higher equilibria, from vicious to virtuous cycles. (It if presumes path dependency in this respect, of course, it also proposes a form of convergence, in that the long-run equilibrium of a system is absolutely vicious levels of inequality. Who exactly is on top, though, is contingent, and continuous shoveling may make the long-run level of inequality other than its in-principle equilibrium one.)
The ontological implication, which flows directly from the strategic one, is that the wretchedness or glory of any particular group does not have to be attributed to their essential characteristics, but can be understood as historically accidental and therefore changeable. If an important category of conservative futility arguments are arguments about group essences – that blacks are inherently stupid, violent, and lazy, or women inherently submissive and illogical – then in its ontological mode the feedback thesis is not just anti-futilitarian but specifically anti-essentialist.
The epistemological implication – perhaps even deserving its own subcategory, an “ideology thesis” – flows from the fact that cultural power and dominance in the realm of ideas are also forms of power that can be purchased with other forms. The classic statement of ideology is, of course, Marx’s epigram that “the ruling ideas of every age are the ideas of its ruling class.” In historical materialism, of course, the claim is that conversion in one direction – from economic to ideological power – has fundamental explanatory power. But regardless of this prioritization, the ideology thesis has powerful implications: that the entire field of available knowledge is biased “to the right,” that is, to those who hold greater power in various regards. Almost all the leftist rhetorical impulses that opponents find maddening enough to give pride of place in their parodic accounts of that rhetoric – from the dismissal of rival economic theories as “bourgeois” to invocations of false consciousness – have their root, to the extent that they are really expressed, in the inescapable logic of the ideology thesis. (A less epistemologically paranoid, but still generalizable, form of this concerns the deduction since power in general produces the power to structure public policy and social structure more generally, the structure of society is “to the right” of where it would (should) otherwise be – and that we may be relatively sure of this regardless of what form that structure takes.)
Rightist appropriations of the feeback thesis can take one of two forms, either of which tends to be an appropriation of the ideology thesis in particular. The first form is its application to cases where, for reasons of historical accident, the right can make claims, basically similar in structure and implication to the left’s, that it speaks for the marginalized and powerless, such as the rural or unborn. (The pro-life activist, after all, is hardly making an implausible claim when she notes that the unborn cannot advance their own interests.) By necessity, these tend to be rather domain-specific. The broader meta-ideological counterclaim concerns the right’s own basis for epistemological paranoia – that the left is in fact in strategic control of the means of ideological production, having the majority of schoolteachers, journalists, academics, artists, and intellectuals more generally in its pocket. Invocations of “the liberal media” may appear irritable mental gestures masquerading as ideas, but are actually coherent arguments against everything supposed to be held by that media. (“Neoreactionaries,” who expand the folk concept of “liberal media” into the more theoretically elaborate “Cathedral,” carry this epistemological paranoia to its fullest logical conclusions.) Leftist responses to this, in their turn, can either dispute the major premise – noting, for instance, that these ideological state apparati are ultimately still opposed to the abolition of capitalism, race, gender, the family, and so on – or the implicit one – noting that ideas can be produced by truth as well as power, and thus that the relevant consideration is not the distribution of ideological power per se but the degree to which it conforms or fails to conform to the overall distribution of other forms.
II. The Induction Thesis
The induction thesis is relatively straightforward. While perversity, futility, and jeopardy all claim that some proposed change will have consequences that are baleful at best, the induction thesis retorts that, in fact, most changes have turned out fine. Almost every major change that conservatives at the time screamed about has turned out quite fine. Getting rid of kings didn’t lead to anarchy. Women’s suffrage didn’t lead to shrews conscripting hapless men into marriage (as hilarious as that might have been.) Just generally we’ve had two to three centuries of things getting better and betterer as they’ve gotten lefter and lefterer.
In this, the induction thesis on its lonesome simultaneously blows the whole conservative troika out of the water. Whatever the real effects perversity, futility, or jeopardy might point to, and even if we weren’t able to articulate any countervailing effects, those effects still obviously exist and outweigh the perverse, futile, or jeopardizing ones. To this spectacular triumph we must append three important caveats.
The first is that the major premise can be disputed. Did things get better? Moderate conservatives typically claim that things have gotten better, but then gotten worse in their lifetimes – endorsing first but not second-wave feminism, or emancipation but not desegregation, and so on. They might well be right in this, but the induction thesis is a powerful argument that they are unlikely to be. But radical antimodernists can coherently claim that life has gotten worse in important respects, and for a long time, too. We may have accumulated more toys, but society has become more atomized and less spiritual, the elimination of ties of dependence has deprived us of important virtues, and the physical environment has been ravaged. (It is far from unusual, and in fact it is probably the rule, for rightists and conservatives who are not libertarians or fusionists to be sincere environmentalists.[3])
The second caveat is that it is agnostic about the actual reasons things have gotten better. It is in principle conceivable, even if most people would not credit it with tremendous plausibility, that although advances in the physical sciences have tremendously improved our quality of life, we would enjoy these fruits //even more// if we had retained the social institutions of 1066. As far as I know no one advances that argument explicitly – even making the implicit explicit is a popular way to make fun of anarcho-capitalism – but more serious claims are advanced on behalf of 1776, 1830, 1930, and so on. Moderation on this front is open to the same weaknesses as noted above, but regardless, one can coherently credit undeniable advances to factors other than changes in social institutions, such that the social changes can be held to be baleful in effect.
The third caveat is that this is only true in the aggregate. Any particular proposal might still have terrible effects. Nobody today claims that Prohibition worked, even if they whistle and look the other way when discussing modern drug policy. Nobody other than friendless weirdos like me[4] thinks Communism worked great. And even I’m reconstructed enough a Stalinist to admit that plenty of Communist initiatives, like collective farming, turned out to be major mistakes.
Implicit, then, in most versions the induction thesis is that while the left should win most of the battles, it shouldn’t win all of the battles. Radicals can use induction against liberals to argue that “imprudent” revolutionary change has a good record, of course – the entire rhetorical value of the concept of “bourgeois revolution” lies precisely in this, and why radicals tend to argue against liberals not just that the violence necessary to establish capitalism was great (which, on its own, could be a simple mote in the eye of capitalism and liberalism) but that it was necessary (establishing a precedent for revolutionary action.) But by default, the mode lends itself to cautious optimism.
Hirschman noted that if the futility thesis is substantively more “moderate” than the perversity thesis, it is not necessarily more respectful, and indeed can take the form of vicious mockery. This is doubly true for the induction thesis. In terms of its expression, the most remarkable thing about the ways it is invoked is how frequently it is mean, humorous, and “speaks for itself” through the simple quotational use of older conservative opposition to positions now accepted by even mainstream conservatives. In the age of social media, these have achieved easy virality.

Like the appropriations of feedback that do not embrace epistemological paranoia, right appropriations of the induction thesis tend to be domain-specific for obvious reasons. This may be the rarest of all six possible forms of appropriation; in fact, I can’t think of many examples other than outside of the domain of arguments for free markets and technology against protectionism and calls to restrain labor-saving technical innovations, where there is a record of people on the left making disastrous predictions that were, in retrospect, inaccurate.

There was a Mallard Filmore cartoon that used a simple induction argument – interviewing some smarmy moron liberal and pointing out how the Laffer Curve has always been right – but… well, I can’t find it immediately on Google, and while I could wade through a thousand Mallard Filmores… I’m… just not going to do that. I’m good. Thanks.
III. The Expertise Thesis
Whereas the induction thesis is agnostic about the mechanisms behind “progress” – that is, why futures tend to be better than pasts, all things considered – the expertise thesis rests on a specific mechanism – that the arrow of social time is deducible from the arrrow of epistemological time. The future knows the past, the past doesn’t know the future, and therein lies the difference. We know more than we used to, and so our judgments are better than those of the past.
Whereas the induction thesis mocks the past, the expertise thesis looks down upon it. We are lucky enough to live in more enlightened times – though not, presumably, lucky enough to live in even more enlightened times, as our descendants will.
Two caveats are necessary in speaking about the expertise thesis, which, together, make it probably the most appropriated of any of the six. The first is that while is it gnostic about mechanisms, it is agnostic about content (as opposed to the induction thesis, which can point to the leveling character of social changes.) The second is that its logic not only speaks to the relationship between future and past, but between more versus less educated groups of people. “Educated” here needs not refer to formal schooling; standpoint epistemology, for instance, is an attempt to theorize the oppressed as fundamentally more capable of viewing reality as it is than are their oppressors, who more fully mystify themselves through ideology.
As with appropriations of the feedback thesis, appropriations of expertise can be more domain-specific or quite radical. The less ambitious, domain-specific appropriations seek to establish that social scientific (or sometimes life-scientific) advances, or perhaps just long-standing consensus among experts in the relevant field, supports the right-wing policy under discussion. Contemporary conservatives can justly point, for example, to the consensus against tariffs and trade restrictions among professional economists. Racists both in the 1790s, 1890s, and 1990s believed that recent advances in biology had justified their position.
It is characteristic of the expertise thesis both in its “natural” and appropriated form that opponents are metaphorically constructed as religious dogmatists. Since the Enlightenment, “the Church” has been constructed as a case of backward darkness and superstition, and it is remarkable that even explicit criticisms of Enlightenment ideals themselves have tended to take this form since the 1890s. If the doctrine that no men should have sex with each other is constructed as a basically idiotic, religious injunction, so is the doctrine that they are all created equal. Consider eugenicist Lothrop Stoddard:
The fact is that modern science is unquestionably bringing the democratic dogma under review. And it is high time that scientists said so frankly. Nothing would be more laughable, if it were not so pathetic, than the way scientists interlard their waitings (which clearly imply criticism of the democratic philosophy) with asides like: “Of course, this isn’t really against democracy, you know.”
Now these little pinches of incense cast upon the democratic altar may keep near-heretics in good standing. But it is unworthy of the scientific spirit, and (what is more important) it seriously retards progress…
When, therefore, believers in race betterment are accused of being “undemocratic,” they should answer: “Right you are! Science, especially biology, has disclosed the falsity of certain ideas like ‘natural equality,’ and the omnipotence of environment, on which the democratic concept is largely based. We aim to take the sound elements in both the traditional democratic and aristocratic philosophies and combine them in a higher synthesis — a new philosophy worthy of the race and the civilization that we visualize.” [5]
The more generally applicable rightist appropriation of the expertise thesis is that expertise is distributed unequally, and those with the expertise to have climbed to the top of it are those who should. In its radical rightist implications, it is similar to the radical leftist ones of the feedback thesis, fully considered.
An important subtype of the expertise thesis might be called the adaptation thesis. In what might well be considered its own class of argument, were I not arbitrarily attempting to squish these into three left theses to match Hirschman’s three right ones, this proposes that a change to society is necessary because society itself has changed. However, in three important respects, this is basically similar to the expertise thesis. First, it empowers experts, because it generally requires expertise to recognize. Second, its basic opposition is between backwards and modern. Third, it is formally neutral in its content, and thus lends itself easily to appropriation.
IV. Tensions and Contradictions
Hirschman noted that the three conservative theses exist in tension with one another; the futility thesis, for example, assumes that reality is basically static and unchangeable, whereas the perversity thesis assumes precisely the opposite.
Of course feedback, induction, and expertise – like perversity, futility, and jeopardy – are hardly entirely in conflict, and share resonances as well. After all, each trinity tends to be invoked by the same groups of people, and not merely (one hopes) for reasons of expediency. Positive feedback loops and the asymmetric nature of accumulating expertise over time can be understood to form the basis of inductively apparent progress. Ideological mystification produced through feedback can at once exist and strain under the weight of increasing societal expertise. That increasing societal expertise can in part be a result of our increasing awareness of feedback, and of the increasingly large sample size of the induction.
But the tensions are real. Accepted on their own terms, they imply very different understandings of history and of proper political action.
On its own, for instance, the feedback thesis actually implies that things should get – from an egalitarian perspective – worse and worse, not better and better. When Karl Marx expected exploitation to increase with productive capacity not just over the history of capitalism but over the history of class society as a whole, he was only explicating the basic logic of the feedback thesis. Marx was hard-headed enough to understand by “progress” that increase in productive capacity. But on the inductive view, the feedback thesis must be as generally wrong as the perversity, futility, and jeopardy theses – or, at least, since those are about outcomes and feedback is about a mechanisms, it must be overridden by other factors even if it is not “wrong” as they are. For its part, the expertise thesis argues that things get better over time as surely as the induction thesis presumes it.
If they are not equal in their commitment to progress, they are likewise unequal in their commitment to equality. From the perspective of feedback, any egalitarian transformation of society is always justified. Of course one cannot deduce simply from feedback that we should all live in communes, but feedback presumes its own limitations on effective action – that power will always accumulate to those who have more power, over and above what a Rawlsian minimax principle (or whatever) would allow. Reality is always to the right of where it should be. Induction, as noted, more weakly can say that egalitarian social change has been good – at the very least, has not been so bad as to collapse the allegedly fragile social fabric that conservatives insist we find ourselves in. While feedback arguments tend to imply radicalism, inductive arguments tend to imply liberalism, in that the ideal political framework would be one where the left wins most, but not all, of the battles – as left-liberals claim is the case for the current social order. For its own part, while inherently pro-change, expertise and its little sister adaptation are indifferent to the matter. Indeed, expertise in its naive form tends to be an argument for the class interests of the intelligentsia.
This – along with the perhaps more obvious matter of the secular increase in the power of experts, bureaucrats, and managers over time – may explain why skepticism towards progress and skepticism towards this social group tend to go hand in hand, and not only among ordinary conservatives. Indeed, the logic of the feedback thesis may lead one towards an unconventional social conservatism, insofar as “social progress” on the cultural front – the dismantling of traditional, local-level systems of authority such as the patriarchal family – can be understood as moves by which the ascendant managerial class cuts off the tall poppies that might stand in its way. It is by this sort of class analysis that Marxists like Alasdair MacIntyre, Christopher Lasch, and Eugene Genovese have been won over to conservatism without sacrificing their historical materialism. Many anarchists a weaker form of these arguments, where if they do not reject feminism and other cultural corollaries of the Enlightenment, they do express concern that the rest of the left – in either its social-democratic or Leninist forms – has been captured by the interests of paper-pushers as a class, or simply by “the state” in abstracto.[6]
An additional feature of their varying attitudes towards equality concerns the fact that feedback is interested in relative outcomes and the distribution of power among different groups and individuals, whereas induction and expertise take it for granted that there is an unproblematic common good. Part of this again speaks to the fact that experts may speak to their own sectional interests, representing it (as all ruling classes do, the ideology thesis would have it) as that of the whole society.
Beyond this, an additional irony – and argument that liberals may fairly raise against radicals – is that each of these arguments, like perversity and jeopardy, is weaker the more commonly accepted it is. In principle, we might overcorrect for feedback effects, start taking the progressive nature of social change for granted, and take for granted as well our superior epistemic position to the past, throwing out the “congealed wisdom” whose inheritance would otherwise actually guarantee it. The smarter and more abstract conservative arguments are to the effect that this has already happened.
V. Parting Thoughts
Finally, some words on the assumptions embedded in this schema itself.
First, the parallelism here assumes that the relevant distinction is between left and right, not between (say) right, liberal, and left. Partly this is because it is constructed in response to Hirschman, who understood his trinity as a set of arguments that are invoked both by radical rightists as well as right-liberals. For the purposes at hand, I think, this makes sense, because they are primarily not arguments that are made in constructing an ideal static model of society, but in favor of moving society “to the left” or “to the right.” In this respect, we might expect a classical liberal to invoke “left” forms of argument in the 1770s and “right” forms in the 1970s, even though his substantive vision of the proper social order remains relatively unchanged.
Second, although all of Hirschman’s trinity have been levied against egalitarian social change, it is change, not egalitarianism, they are concerned with. In this respect, I have not quite kept the parallelism, except, if one wishes to stretch, in that feedback effects assume that the need for change will be constant if one wishes to preserve equality.
Third, although I perhaps only folded in adaptation to expertise for the sake of a parallel number of arguments or theses, I have not constructed these to parallel Hirschman’s three individually. I do not think that a set of direct, polarity-reversed parallels would be particularly interesting.[7]

Last, I don’t pretend that any of the individual insights here are original (even where I can’t recall the source.) If this sort of thing has any value, it will be in a schema itself achieving fixation as a set of common terms to categorize arguments – as perversity, futility, and jeopardy have. The above is, I imagine, inadequate to that task. But as there is no widely acknowledged parallel to the original schema
VI. Notes
[1]In fact, this is obvious enough that I’m certain it’s been done before; indeed, as I write this, I haven’t even finished Hirschman’s book, so maybe he comes up with left complements himself! So that’s certainly something I should look into. But there isn’t any counterpart set that’s as well-known as the original PJF trinity, and in any event, I suspect composing this prior to exposure to others’ attempts will offer the possibility of some amount of convergent validity, if there’s some real virtue here.
[2] Bourdieu, Pierre. “The Forms of Capital”
[3] The logic of this actually extends further than the simple time series regression that we’ve mastered nature more and more as progress marches forward. Burkean arguments for the fragility of complex systems and the perversity of artifice are basically homologous to the major environmentalist arguments, and absent further arguments to distinguish them, if you accept one you should accept the other.
[4] And the majority of the FSU, depending on which poll you trust, but that’s another story.
[5] Lothrop Stoddard, The Revolt Against Civilization: the Rise of the Under-Man, pp. 265-66. They don’t title books like they used to!
[6] I have written a bit on the longue duree of left, right, and bureaucracy here, if you’re into that sort of thing.
[7] After having written most of this, I now see that it is what in fact Hirschman proposed. This may be, perhaps, why they have not reached fixity.