Why the left is right

A philosophical defence of modern partisanship

A while ago, I wrote this draft argument with the plan of eventually submitting it to a journal- a project I am no longer interested in, as I have left academia- so I thought I’d publish it here. Beware, it is not fully complete.

Essentially, I argue that we have reason to think the left is correct about most issues, contra the work of a previous author, Hrishikesh Joshi, who argues that it would be implausible if one side of politics were right about a large majority of issues.

A defence of the left wing of politics and an argument against Joshi

Joshi argues in rough terms that it would be extremely unlikely for one side or the other to be correct about a majority of political issues because politics consists in a bundle of apparently separate questions, and it is unlikely that one side or the other is right about all or most of them. I respond by identifying a trait which might explain both the formation of political beliefs, and provide an explanation for one side of politics may be more likely to maximise desire satisfaction. Social dominance orientation seems to form a strong component of right-wing political beliefs, but we would expect policies formed through a social-dominance orientation outlook to not serve the general welfare. I conclude by considering additional questions raised by the left-right division in philosophy.

Since you know as well as we do the right, as the world goes, is only in question between equal power, while the strong do what they can and the weak suffer what they must.

-The Athenians, quoted in the Melian Dialogue, Thuycidides circa 404 BC

Joshi’s argument and the roots of our response

Joshi summarises his argument as follows:

People’s political beliefs cluster around two main camps. However, many of the issues with respect to which these two camps disagree seem to be rationally orthogonal. This feature raises an epistemic challenge for the political partisan. If she is justified in consistently adopting the party line, it must be true that her side is reliable on the issues that are the subject of disagreements. It would then follow that the other side is anti-reliable with respect to a host of orthogonal political issues. Yet, it is difficult to find a psychologically plausible explanation for why one side would get things reliably wrong with respect to a wide range of orthogonal issues.

The core of our response to Joshi will be that political issues may not be orthogonal. We aim to motivate as plausible the following theory:

1. Different political opinions may reflect a common pattern of preferences for prioritising either the powerful or the powerless.

2. There is a systematic tendency for power and powerlessness to correlate with political rightness or wrongness relative to certain moral standards that many will agree are likely to be approximately accurate.

3. Thus, there are reasons to think one side or the other is more likely to be right, or at any rate righter than their opponents.

We cannot, in this paper, show that our theory is true- we cannot even show that it is likely to be true, we merely aim to show that it is, in Joshi’s words, “a psychologically plausible explanation for why one side would get things reliably wrong”. Yet our goal is not to quibble with Joshi for the sake of quibbling but to open up ways of thinking about the categories ‘left’ and ‘right’ that have been largely absent from political philosophy, social epistemology, and analytic philosophy generally.

We will use the word “chirality” to mean a side of politics- the left-wing or the right-wing. Although there are sceptics of the idea that politics can be split into a left and right side, we will follow Joshi in assuming the split is coherent in a sociological sense- i.e. that the concepts of “left” and “right” are useful for predicting the policy positions of ordinary people, activists and politicians. This in no way presupposes that the left or right represent coherent positions, or that believing in some left or right positions gives you reason to believe in others, or that there are not even more precise and informative categories that would allow us to predict with greater accuracy what people believe about politics.

Analytic Political philosophers, and philosophers in the Analytic tradition generally, do not often explicitly speak of the political left and right. I finish by arguing that the abstraction away from questions of the political left and right by political philosophers is regrettable. Chirality is not an especially popular topic of theorisation in political philosophy, yet it plays a mammoth role in folk political discourse, political organisation, and folk political philosophy. References to clearly left-wing and right-wing ideas are common in philosophy, but not to the organising concepts. The left and right represent spontaneous coalitions of political ideas and philosophies, and engaging with them may well increase the relevance of our work.

A Tentative Argument That The Left is More Correct than the Right About Most Issues

I will now outline an argument that, contra Joshi, the left may be correct about most policy matters. A full treatment would have to be book-length, but we aim to describe the view and give it some motivation here, to show plausibility and initial promise.

But what can right and wrong even mean in the absence of moral common ground?

The question of how moral realism and anti-realism relate to Joshi’s considerations is a difficult one that Joshi himself does not deal with in depth. Although Joshi does discuss expressivism and other forms of non-cognitivism, these are not the only forms of anti-realism. If certain forms of moral anti-realism are true and people on the left and right vary systematically in their values, it may be that both sides can truly assert “I am right about X political issue”. In addition to these problems of ethical and metaethical variance, it is often unclear in political debates how much of the debate really depends on normative or positive considerations. Joshi himself never gives an account of what it means to be right about political problems. Given the amount of disagreement on ethical and meta-ethical questions, it would be useful if we could change the question of which side is more correct from a debate with mixed ethical and positive components to a debate over which view is best, assuming a particular ethical system. I will use a particular ethical view- an ethical view that I do not hold myself- as a yardstick for the remainder of this piece, preference utilitarianism. Somewhat more exactly:

Out of a set of political actions, X should be chosen if and only if X will lead to the greatest degree of (non-instrumental) preference satisfaction averaged over all people.

Naturally, there are many subtleties in the correct formulation of preference satisfaction utilitarianism. There are problems, for example, of interpersonal comparison. But the resolution of these problems will not be, as far as I can tell, essential to any claim in this paper, and we will assume that such issues can be patched up.

Note- the sense of correctness we’re going for here is a comparative sense of correctness- that is we are interested in the question of whether one side is more correct about political questions that divide it from the other side. In this sense, the left can be correct even if both the options argued for by the left and right are greatly surpassed by some third option, perhaps unknown, or perhaps argued for by some small minority.

But even if we manage to show that the left is more correct than the right on a majority of issues, on a preference satisfaction utilitarian account, why care unless one is a preference satisfaction utilitarian? Our results are relevant to those who are not preference satisfaction utilitarians. Firstly, Adler & Posner (1999) and Hausman contend plausibly (2010) that preference satisfaction can be taken as an indicator of welfare, and a strong indicator at that, even if it is not welfare in and of itself. Secondly and relatedly, if your objection is to the selection of preference satisfaction, many other conceptions of human welfare (e.g., the balance of pleasure and pain, flourishing, capabilities) will tend to move in tandem with it. This means that, with modest assumptions, our approach can be extended to other forms of utilitarianism apart from desire-satisfaction utilitarianism. For reasons that will become clear, it can also be extended to forms of welfarism that are more egalitarian than utilitarianism- for example, prioritarianism or a max-min social-welfare function.

But what if you endorse an ethical system that includes side-constraints or goals other than welfare? Here, I would say that we will have at least shown something useful if we can show, or even give evidence, that one political view or another is welfare maximising, as on many non-welfarist ethical views, welfare is still a powerful factor in decision-making.

Our argument is as follows:

1. The left has a strong and broad tendency to favour the interests of the relatively powerless, whereas the right has a strong and broad tendency to favour the interests of the relatively powerful.

2. Generally speaking, the unfulfilled desires of the powerless have greater urgency than the unfulfilled desires of the powerful. Part of what it means to be powerful is the capacity to fulfill one’s own desires.

3. So, from a preference utilitarian perspective, we have good prima facie reasons to favour the unfulfilled desires of the weak.

4. Therefore, from a preference utilitarian perspective, there are good prima facie reasons to think that the left is more correct than the right, because, and insofar as it favours the weak.

The evidence for proposition 1

What is the evidence for Proposition 1? Some people- leftists, and interestingly, even some rightists- upon hearing it articulated, immediately assent to its truth, many, though, will deny it. Is there any scientific evidence?

Social-Dominance Orientation (SDO) is a construct that refers to both a preference for social hierarchy and a preference that one’s own group should be higher in the social hierarchy. Social dominance Orientation strongly predicts voting for right-wing parties and supporting right-wing ideas (Pratto et al., 1994; Pratto et al. 1996; Duriez & Van Hiel 2002; Van Hiel & Mervielde, 2002). These effects are also pronounced -even more so- among party activists, in an Italian sample, Mebane et al. (2020) found that centre-left activists had an average mean SDO score of (1.84 if male, 1.79 if female) and centre-right activists had an average mean SDO score of (3.27 if male 3.03 if female) [you need to give some sense of effect size and present the statistics better].

It is worth commenting on the effect size here. Consider, for example, Pratto et al. (1994), which found the following Pearson correlations between measures of policy preferences and measures of Social-Dominance Orientation: support for social programs (r=-0.47), support for gay and lesbian rights (-0.37), anti-black racism (0.55), political economic-conservatism (0.38), and support for environmental programs (-0.38). The real correlation, due to attenuation for the unreliability of measurement on both sides, would very likely be even larger, and the connection appears to extend across virtually every political issue.

Joshi coincidentally discusses effect sizes in social science in the very paper we are critiquing. He mentions work by Cullen (2018), which finds correlations of a similar size between different policy preferences (e.g., one’s stance on immigration and one’s stance on guns). Joshi comments that such relationships are very large in the context of social science:

These correlation coefficients indicate a “moderate relationship,” in statistical parlance, but they are quite striking results in a social scientific context. To put them into perspective, the correlation coefficient between the height and weight of male college students tends to be around 0.45

Nor is this phenomenon simply an upshot of SDO questions that directly test political beliefs. Consider the Short SDO test (Pratto et al. 2013), which, except for length, is representative of SDO tests in general:

1. In setting priorities, we must consider all groups.

2. We should not push for group equality.

3. Group equality should be our ideal.

4. Superior groups should dominate inferior groups.

These questions seem, at their face, to directly assess the topic they claim to assess- the preference for some groups to be dominant over others, and a preference for the power of dominant groups over weak groups. They are not directly questions about left-wing and right-wing policy preferences, except inasmuch as preferences regarding group equality are left-right coded. Nor can these questions all be interpreted as preferences regarding government intervention or non-intervention to equalise groups, see question 4.

These results give evidence for the claim that the right generally prefers the strong over the weak, and given the effect sizes, that preference is substantial.

The biggest weakness in our argument here, and one we cannot address at paper length, is that SDO may be strongly correlated with being right-wing, and with specific right-wing choices, but could be ephemeral to the ideology of the right. Perhaps rightwingers generally have a high social-dominance orientation but it is not a major driver of policy preferences, and further, perhaps carefully examined, rightwing policies have no pattern of favouring the strong over the weak.

To me, and I hope to many readers, this seems implausible, because SDO and related attitudes seem like an active part of right-wing ideas and proposals, but perhaps others may find the separation attractive. A deeper study of the degree to which SDO influences right-wing policy preference formation might include strategies like a qualitative review of right-wing justifications and reasoning, interview studies of political subjects, or some attempt to intervene on SDO as a variable and consider subsequent changes to policy preferences.

Certainly, there is room for argument. Conservatives might claim, for example, to care more about the weak than leftists do in the case of, say, abortion (to which many leftists would respond by denying the personhood, and thus ‘strength’ or ‘weakness’ in the relevant sense of the fetus). Still, the framework is potentially applicable to many issues beyond the obviously ‘economic’: from same sex marriage, to incarceration, to immigration, and a path to citizenship, etc. It does not have to explain every single issue in order to refute Joshi’s thesis.

The evidence for proposition 2 and the inference from 2 to 3

The simplest case of 2 is the economic case. It is commonly thought (Layard et al. 2008, Kahneman & Deaton 2010, Clark et al. 2008) that increases in income have declining marginal effects on welfare. This function is thought to be at least logarithmic and maybe significantly more concave than that. If the function is logarithmic, this means, for example, that an extra dollar is worth twice is much to someone earning 50,000 than someone earning 100,000, and twice as much again to someone earning 25,000. An increase in income, or equivalents like goods, is dollar for dollar more desire-satisfying for the poor than the rich. Hence the political-economic needs of the constituency of the left are more urgent than those of the right.

However, this phenomenon can be found beyond the economic realm. Generally, the unfulfilled demands of the weak will be more urgent than those of the strong, because they have less political power. Consider, for example, a simple model in which the probability of getting what you want politically is a positive increasing function in your power, and of the urgency to you of getting that thing. It’s easy to see that, barring extraordinary coincidences or other unusual factors, the average urgency of the things the powerful want but do not have will be lower than the average urgency of the things the relatively powerless want, but do not have.

We can also think of a model in which part of what politics does is distribute prestige and respect - again, the needs of the powerless for respect will probably be greater than those of the powerful.

Joshi’s views about the weak and the strong

Joshi grapples with reasoning analogous to ours:

Consider for example, Immigration, Wages, and Affirmative Action. One possible answer that is worth considering is the following. Perhaps liberals and conservatives assign a different importance to helping those who are the worst off. Liberals assign a greater importance than conservatives to helping the neediest. Immigration, Wages, and Affirmative Action are thus not orthogonal – they all flow from a basic principle according to which it is especially morally important to help the worst off. So, for instance, the Priority View, defended in Parfit (1997) is an example of a moral theory that could fill this role. According to the Priority View, benefits to individuals matter more morally the less well-off they are. A conservative partisan could enlist this putative observation as a way to explain the anti-reliability of the liberal side; she could claim that since the Priority View is false, and that the views which are to be found within the liberal cluster of opinion are applications of the Priority View given the empirical facts, liberals are prone to systematic error with respect to political issues. And, conversely, the liberal may contend that the conservatives’ anti-reliability is to be explained by their non-acceptance of the Priority View.

Firstly, we note that there is no reason to think that our argument depends on priority at least in the sense of prioritarian ethics. Maximising priority weighted welfare will quite possibly also lead to leftism, but our argument gives reason for a utilitarian to favour the interests of the weak, and the utilitarian does not think the worst off are intrinsically more deserving of extra attention- just that they should be helped first becasue they will in fact benefit most from being helped.

Now, what about the question of empirical rightness or wrongness? Joshi here is effectively making an appeal to the idea that policies intended to help the worst off may be self-thwarting. Such arguments try to block the inference from proposition 2 to proposition 3 by reasoning that while in the abstract the needs of the worst off may be more urgent, in practice, attending to such needs by the left’s proposed means will not be good even for the worst off. These arguments are most well-developed in the sphere of economics. Here, there are many arguments that leaving the market alone- or even actively favouring the interests of the rich using the instruments of policy- will be better for everyone.

However, such arguments have not just been made in the economic sphere. Theorists have argued that even though the right favours the interests of the strong and the left favours the interests of the weak where they compete it is, from a long-term perspective, in the interests of the relatively powerless that the powerful win their fight. Perhaps, for example, for reasons of political stability.

For a classical instantiation of this argument, see Pareto (1916), who argues, in essence, that society is always defined by great inequalities of power, and attempts to correct this will only result in a “circulation of elites”. In the rhetoric of reaction, Hirschman (1991) identifies similar arguments to these as arguments from futility and perversity.

Another form of this argument goes:

1. When the powerful have what they want, a society is robust.

2. The greatest risk to overall welfare in a society is collapse or destruction.

3. Therefore, we should give the powerful what they want.

There is a strong counterargument:

All things considered, the plans and policies of those who care most directly and strongly about the weak, as shown by their low SDO scores, are likely to favour the weak.

Consider a situation in which there are two available trustees who will make different choices on behalf of the object of the trust. One of these trustees clearly seems to care much more about the object of the trust than the other, but the question of whose approach will work better is impossible to answer. It seems like the evidence that one trustee cares about the object of the trust than the other is evidence in favour of that trustee.

Likewise, in an overall sense, it is of course possible that a social group will be harmed by those aiming to raise it up through complex second-order mechanisms, and that the program of the political coalition that cares the least about a group may actually be to their greatest benefit. However, there are strong prima facie reasons to think that the interests of a group are best served by those who value that group and its equality with other groups, not those who think it should be subservient to others. Is such an argument anywhere near certain? No, but it doesn’t have to be, Joshi writes:

All I want to show is that the Priority View, supplemented with particular empirical assumptions that aren’t unreasonable and are up for debate, can lead to the conservative line on racial/economic issues. Moreover, since the different empirical debates are presumably orthogonal, the issues themselves are themselves orthogonal – a particular answer on Immigration does not commit you to a particular answer on Wages or Affirmative Action.

But here Joshi makes an error- it is not actually true that rightness or wrongness on each of these debates is orthogonal- or perhaps it is better to say that they are orthogonal but not in the sense he needs. Suppose that some company and its scientists claim that a product of theirs helps human health in five different ways and some other scientists from outside the company disagree with this view, saying that none help. It could indeed be that, in a sense, the efficacy of each mechanism is orthogonal- but there is also clearly a sense in which the evaluation of the likelihood of each mechanism succeeding is not orthogonal. Although the right and the left give different arguments and propose different mechanisms by which various of their policy proposals might work, it would not be a coincidence if the side that aims at helping the worst off the most does, in fact, help the worst off the most.

Thus, while it is true that on each of these issues, it is possible that the leftwing position is self-thwarting, and to this extent they are orthogonal, but the defender of taking a common leftwing line on each of them has a reason to think they might be right on each. There are strong prima facie reasons to think that the interests of a group are best served by the policies of those who value that group and its equality, not those who think it should be subservient to others. There is thus a path to leftwing reliability, and rightwing anti-reliability on these issues, although it is not certain, and this uncertain conclusion is all we need to reject Joshi’s claim that there is no plausible mechanism by which the left might be correct on a vast range of (apparently) orthogonal issues. We cannot hope to refute all counterarguments to our position in such a short article as this, and so we cannot definitively establish that the left is right about most things, but I hope to have shown the outlines of a path.

An alternative

There are many stories one could tell of the form:

1. People who are high in trait X are more likely to be supporters of political team Y

2. Being high in trait X should be correlated with making the right decisions on political questions from a preference utilitarian perspective

3. Therefore, all else being equal, we would expect that members of Team Y are more likely to be correct on political questions.

The possibility that one of these accounts is true tells against Joshi.

Altruism: Multiple studies have found that leftists are more inclined to impartial altruism than rightists (e.g., put in charity game studies), and there is also some evidence that leftists are more inclined to altruism fullstop- whether impartial or not (Other charity game studies). There are clear theoretical reasons why the more altruistic side of politics might, in general, endorse those views most likely to maximise preference satisfaction.

Personality: Studies have found that leftists score higher on Openness to experience Conservatives score higher on conscientiousness. One can conceivably imagine that policies supported by the open to experience or the conscientious are more likely to be correct, and this could provide basis on which to think either the left or right is more correct. Interestingly, it is not the other-directed components of conscientiousness in which conservatives excel. When we consider honesty-humility as a separate trait, which includes things like fair dealings with others, it is leftism which positively correlates with the personality trait, and the relationship is relatively strong (e.g., r= in ---). Honesty/humility also clearly could be correlated with supporting the correct views.

Many of these stories have a weakness relative to ours, namely that they do not explain what is held in common between different leftwing and rightwing ideas, although at least two immediately suggest such a story- differences in altruism, and differences in consequentialism versus non-consequentialism. Many other connections are possible, We note that several of these stories have possible overlaps with our own. Most notably, there is an obvious confluence between a story based on support for the weak and a story based on altruism, or for that matter, honesty/humility.

Political Philosophy and Chirality

What lies beyond refuting Joshi’s argument? As previously mentioned, philosophers often abstract away from concepts such as ‘left-wing’ and ‘right-wing’ in their work- yet these are the grooves on which real politics moves, and an engaged political philosophy should have much to say about both the essence and correctness of each. We have given a theory of one of the important differences between each chirality, and explained how this might relate to the correctness or incorrectness of each. We hope that in the future, questions of political chirality and the divisions and differing claims of the left and right might become more central to political philosophy; they are, after all, central to how almost everyone except political philosophers thinks about politics.

There are at least four questions of political philosophy in relation to questions of chirality. Political philosophers cannot claim exclusive ownership over these questions -other disciplines should have much to say about them as well- but they are very clearly in the ambit of political philosophy.

1. The content question: Are left and right ‘grab-bags’ of political positions, or do one or both have a common, unifying structure of ideas and/or values? Certainly, this question is partly a sociological question, but the philosopher’s expertise in conceptual and normative analysis may provide assistance here.

2. The reconstructive question: Given the answer to question 1, how should we reconstruct left and right as coherent normative and theoretical frameworks in political philosophy?

3. The connection question: How can we imagine existing approaches to political philosophy in terms of the left and right, in terms of what we find about the content question and the justificatory question?

4. The justificatory question: Given the answer to the content question, and perhaps also the reconstructive question, can we show one political side or the other is more justified or correct about political issues in general, or some significant subset thereof.

The relationship between folk ideas and philosophy is, of course, always controversial. Yet folk ideas have a particular pertinence in political philosophy that they may not have in, say, metaphysics or even ethics, not only because just because they could be right but because they form the substance of the thing being studied- ideas about politics are politics. And also because inasumuch as the legitimacy of democracy creates a connection between how society should be and how people think it should be, the ideas of ordinary people have a direct weight. These questions then, are worthy of further consideration.

Conclusions

We have argued, contra Joshi, that there is a candidate explanation for why one side of politics might be correct on many issues, and the other side correct on far fewer. Given an assessment criteria like maximising the fulfilment of preferences, and empirical results about the relationship between Social Dominance Orientation and Right-wing beliefs, it is plausible that leftwing beliefs, because they are not influenced by social dominance orientation, are more likely to be correct- although we cannot establish that as a certainty here. We listed many other traits which could, at least in principle, explain why one side or the other of politics is likely to be correct about more questions than the other Finally we have called for additional study of the philosophical issues raised by concepts “left” and “right”- and pointed to justificatory, descriptive, connective and conceptual questions raised by political chirality.

I spend many hours a week on this blog and make it available for free. I would like to be able to spend even more time on it, but that takes money. Your paid subscription and help getting the word out would be greatly appreciated. A big thanks to my paid subscribers and those who share the blog around.