Michael Lynch: On Truth in Politics: Why Democracy Demands It

Michael Patrick Lynch, On Truth in Politics: Why Democracy Demands It (Princeton University Press, 2025)

To be honest, I am somewhat discomfited writing a review of On Truth in Politics. There is a lot I agree with in the book, and I am sympathetic with the general attitude of disgust expressed by its author toward the willful expressions of mistruth by politicians and their supporters–particularly by certain ones currently in power in the U.S. I also entirely concur with the author, a Professor and Administrator at UCONN, that anything like an attitude of indifference toward whether political statements are true can be deeply, even irreparably, harmful to a populace. In a word, Prof. Lynch seems to me to be an ally politically, someone whose heart is clearly in the right place. I won’t even deny that his book may be beneficial to some of its readers–though I doubt there will be many of them who didn’t already believe in the importance of truth for good public policy outcomes. 

Nevertheless, it is hard for me to imagine another book containing so many assertions I agree with that is grounded by a thesis that I consider so completely wrongheaded about the natures of both truth and democracy. I will try to illustrate this by first juxtaposing a list of views expressed in the work that seem to me not just true, but important, with another batch of statements that strike me as not just erroneous, but inclusive of a number of items that may even be destructive to democratic principles.

First a bunch of the true ones:

• True propositions ought to be valued more than false ones.
• The belief that some propositions are true and others false need not be inconsistent with pluralism or the freedom to express the wrong ones.
• Facts about things like climate, disease, and history are independent of anybody’s political preferences.
• It is generally quite valuable to a citizenry to have their public policies reflect ideas that are correct; incorrect ones are likely to make them less happy.
• Respect for real knowledge is (therefore) crucial to protect science and history from being gutted.
• “Truth can’t take care of itself when conspiracy, denialism, and the political ideologies that promote them run rampant.” [p. 8]
• The less some group cares about truth, the more harmful to them its public policy is likely to be in the long run.
• That most people–even everyone–LIKE some proposition isn’t the same thing as that proposition being TRUE.
• Democracies need to allow for the ability to safely “speak truth to power.” 

Those are claims that I think everyone ought to agree with, though  I also think that, with the exception of a few “alternative fact” fans of Kellyanne Conway (and maybe some philosophy students who either went to grad school in Pittsburgh or have too much admiration for Richard Rorty wherever they may have studied), a vast majority would assent to each one. After all, who would ever suggest that, even if political pronouncements proceeding from their judgments are exceedingly likeable, the ONLY thing those pronouncements have going for them is that likeability?

OK, so where do I think Lynch has gone wrong? Here are a few of the quite large number of remarks in the book that, partly as a result of his abandonment of a simple correspondence theory of truth, seem to me both to reflect confusion and invite serious defects into any theory of democracy:

• “Political propositions” are true in some sort of different way than propositions about climate, disease, or history, because the latter are “independent of humans in a way that political truth never can be.” [p. 6]
• Democratic theorists and other political philosophers need to share something like Lynch’s theory of truth, or they will never be able to understand how (humanity-infected) political judgments can even BE true.
• No political system can be democratic that does not “value truth.” 
• In a land where likeability replaces truth as the most important determinant of “political correctness,” it’s not clear that anybody can be said to actually believe anything at all. 
• Any good definition of “democratic politics” requires that democracies practice “respectful deliberation” and thus be “spaces of reasons.” [p. 15]
• No jurisdiction can be a democracy unless collective deliberation therein helps us both to do better than we have done before and to arrive at “just solutions to societal problems.” [p. 16]

How did Lynch come to hold both of these packages–and why do I endorse one and not the other? My general diagnosis with respect to the first of those questions is that, in common with several British Hegelians, Lynch has pumped a number of extraneous properties that he happens to find agreeable into concepts that should not be thought to contain them. So, for example, while there is a substantial literature regarding the extent to which deliberation of various kinds can benefit democratic practices, Lynch simply bypasses the entire matter by making deliberation essential to anything that can be called a democratic practice. He does much the same thing with deliberation itself, by pumping into its nature a certain attitude toward truth. On this view, nothing can correctly be called “deliberation” unless it aims at finding truths and so utilizes belief by all parties, rather than any mere non-epistemic commitment to propositions.

To be fair, Lynch does refer once or twice to what he calls “formal democracy,” describing that as something that involves making policy decisions “by voting,” but even if that remark were refined (as it would need to be because, e.g., not any sort of production of policy “by voting” is necessarily democratic–formal or otherwise), what it would give us be would actually be a definition of democracy itself, NOT just of “formal democracy.” By throwing “formal” in there, Lynch suggests that only procedures that he endorses can really be democratic. The problem with such a move is that it implies both that formal democracies might not be real and that real democracies might not require plebiscitary policymaking. 

Now it might be thought that a population with little interest in truth would also not care too much about having or maintaining a plebiscitary system. While I don’t think such a causal argument is quite right (a group like MAGA might decide it was most likely to have maximum power when there are occasional elections that don’t have much effect), it does have some plausibility. As far as I can tell, however, Lynch is not interested in making such a causal argument for the involvement of truth in the maintenance of “formal democracy.” Rather, his goal is to insist that, whatever might seem to be going on, it’s not real democracy unless the electorate is aiming to “get things right.”

This conclusion derives not only from the above-mentioned Hegelian-style concept enlargement, but from a sincere commitment to retain some sort of pragmatic theory of truth,* in spite of the immense difficulties in making any such theory consistent with the clear truth that democracy need not even attempt to produce polities that are thoughtful, happy and just, never mind succeed at doing so. So much of what I have written over the past decade has been focused on this fact which seems to me to be obvious but is lost on many theorists: A legitimately democratic polity is likely to be a cruel or stupid polity if its citizenry is cruel or stupid. That is because democracy is, at its heart, nothing more or less than self-government, a scheme that requires equally-treated people and their equally-weighted choices to be in charge of policymaking. This seems to me obvious, but, as Lynch’s book should make clear, it is quite difficult to square it with any theory of truth that is not a simple, Tarskian matter of correspondence. 

Half aware of this, Lynch strives mightily to somehow reduce the pragmatic element in his theory, but the whole matter is quite easily handled by mucking with neither the ordinary concepts of either democracy or truth. One is right to oppose the contribution of people he calls “Twitbookians” (people of the type who claim Trump is defending our freedom and our democracy when he criminalizes flag burning in spite of a SCOTUS decision). It can hardly be doubted that such folk, caring more about party than truth, are sure to make nearly everybody’s life worse. Surely, if global warming is a fact, denying it because your party leaders prefer it not to exist, endangers everyone. But Twitbookians may be just as fond of democracy as Lynch and I are (I mean, they generally aren’t, but they could be). That’s because (i) democracy is just a matter of people getting what they want, (ii) deliberation is a matter of people discussing what they might want (even, arguably, in the various stupid ways that they prefer), and (iii) truth is just a matter of correspondence with the world.

It should be noted that Lynch sounds almost exactly like a correspondence theorist when he talks about empirical claims.# A statement like “It’s raining in London right now” is either true or false, he says, regardless of what anybody believes or the evidence they have. But “The last U.S. Presidential election was rigged” is thought to be different because it’s “political,” and, on Lynch’s view, political claims, being somehow tied up with humanity in a way that London weather is not, need to be understood differently. 

I believe that’s a mistake. “Political statements,” like “Taxes should be lower” may be taken to be either moral or prudential claims. If the former, whether one believes that one of them is true will depend in part on one’s moral theory (objectivist, relativist, skeptic, divine command advocate, etc.), and it will also depend on one’s confidence in the available evidence. If the statement is thought to be a prudential claim, belief in its truth will be a function of  the effect of taxes on the economy, on thrift, on estimates of future probabilities, etc. In my own view, it’s unlikely that very many people are likely to have sufficient information for anyone to have a great deal of confidence in such claims, regardless of whether they are expressions of prudential or moral judgments. But whatever the propositions being expressed, they are often believed, and will be true or false, depending on how the world is, just like weather claims. An election was rigged or it wasn’t, and higher taxes on the rich are good (morally) or good for us (prudentially–over some long or short term) or they aren’t. Arendt notwithstanding, there is no ontological difference between ‘political’ and ‘non-political’ claims. The sort of evidence needed is likely to vary, but that’s true of the individual members within the groups of both “political” and “non-political” claims as well. 

Lynch notes that both philosophers and political activists, particularly among the left, have often been reluctant to call any “political statements” true, because such claims seem to them to push a Platonic know-it-all attitude. The idea is, if some political positions are correct and the people that disagree with them just wrong, we need an epistocracy, a group of trained “Guardians” to run things for us. The mistake there is again seems to me to at least partly be a matter of confusing ontology with epistemology. While I think I am considerably more hostile to epistemic democracy and its conception of “policy truths” than Lynch is,⧫ there are two important differences in our approaches to such theories. First, I don’t think the relevant error made by epistemic democrats should be thought to extend to statements like “the last election was rigged.” Second, when we move to items like “Jones would be better than Smith as a representative” it is my view that if we disambiguate such claims, they will be found either to signify some person’s or group’s preference (and those are surely not only true or false, but also relatively easy to determine by asking them); or they will instead refer to moral or prudential judgments which, as indicated, are also true or false, but may be exceedingly difficult–even perhaps impossible–to know. 

It is obvious that if an abundance of knowledge is generally good for getting a person what she wants, in a democracy where everybody is regularly asked for their opinion and those views must be acted upon by government officials, the appropriate attitudes toward evidence will usually be beneficial there. I have no wish to deny that. But consider the possible variations. In a complete anarchy, with no government at all, knowledge is surely helpful to members of the populace. In a monarchy, like that of King Arthur’s, where, at least in Mark Twain’s version, people were generally left to their own devices by the state, a Connecticut Yankee’s know-how is again likely to be quite useful. It’s also true that in a completely totalitarian hellscape knowledge doesn’t do Winston Smith any good–in fact it is likely harmful to his happiness quotient. These situations can all be made compatible with Lynch’s theses.

But it seems to be the case too that in a completely democratic jurisdiction where all anybody wants is soma, a lot of knowledge among the masses will provide no obvious prudential gains. And, finally, consider a land run by a mostly benevolent dictator with no pretensions to democracy who asks randomly chosen individuals in her realm for policy advice–and uses it maybe three-fifths of the time. Here again, widespread dispersion of knowledge would likely improve the lot of the citizenry, but the place can hardly be correctly described as “ self-governing.” 

The moral is that Lynch is right that prudential values can be expected to increase in authentic democracies where there is an abundance of true beliefs, simply because in those places, citizens are more likely to make sensible choices which will be required to be used. If that rather obvious remark were all that Lynch meant by the claim that “truth is a democratic value” it would be hard to disagree with, but Lynch insists that he means much more. He claims that there can’t even BE democracy where the search for knowledge isn’t given the highest priority. As he correctly points out it’s not quite that truth is somehow like money, it is that epistemic value generally provides prudential benefits to individuals. So, of course, to the extent that people are given authority over not just their private activities but public policy writ large, a wide dispersion of knowledge can be expected to make things better for everybody. None of that should be controversial; in fact, I believe it would even receive the assent of his “Twitbookians” (those who, in their hearts, prefer “likes” to getting things right). But widespread knowledge is neither necessary nor sufficient for democracy. And it is likely to provide benefits in jurisdictions of all kinds.

It is clearly important to Lynch to retain some remnant of a pragmatic theory of truth, if not for all claims, at least for the so-called political ones. This interweaving of warrant and factuality, of ratio cognoscendi and ratio essendi seems to me bound to make trouble, but, of course, he is not alone in taking that tack. He follows Peirce (and some versions of Putnam) in holding that what is true is, in the end, a function of warranted assertability. FWIW, my own view is that the best response to all such theories is to recommend spending a moment to consider whether it is possible that some proposition that seems true even at the very end of human inquiry, after applying all manner of expert science, epistemology, and all other modes of human understanding to their fullest, is nevertheless just NOT TRUE. All who join me in thinking that this is a legitimate possibility, should believe that pragmatism is false. And, skeptical as I am of human capabilities in pretty much every arena, I have no problem whatsoever in endorsing the realist position. In fact, I don’t believe that any important improvements can be made to Tarski’s definition of truth, which I take to reflect a simple correspondence theory. According to that theory, a statement (or belief, etc.) is true if and only if it corresponds with the real world. And the point is, it doesn’t matter whether we do or ever can KNOW (or even have decent reasons for believing) that some proposition is true or false. It may very well be that for the vast majority of them, we can’t. Alas. In such a world, humility must reign.

A book review probably isn’t the place to engage in a deep discussion of the theory of truth (or even a lengthy, shallow one of the kind of which I am capable). Furthermore, since I am unlikely to convince anybody who already has a view on this subject, I will just say here that the view that metaphysical realism requires that there be a “God’s-eye view of the world” or requires that there be “One true political story of the world” seems to me a confusion, because a “view,” “God’s-eye” or otherwise is an epistemic concept. If, as Lynch says, the fact that the bell has rung “seems like just another way to say it is true that the bell has rung” that is just because Tarski was right. Indeed, I deny that it is the job of a correspondence theory to “explain how plainly political judgments are true” any more than it is its job to explain how plainly scientific judgments are true. What matters is whether what makes them true is that they reflect the way the world is. 

That people disagree about what some society “ought to do” is no more a successful criticism of the correspondence theory of truth than that physicists disagree about the interpretation of quantum “entanglement.” Where there are facts and interested people, there is quite likely to be disagreement, whatever one takes to be the nature of those facts. I am grateful that language somehow allows us to communicate our agreements and disagreements. 

Anyhow, I’ve said enough. In my (most humble!) opinion, this is all we know and all we need to know: (1) Truth is neither more nor less than a matter of correspondence with reality. Statements–”political” or not–have truth values that are a strict function of what properties are exemplified by the world to which those statements refer. 

Again, we may need to figure out precisely what various statements are intended to express before we can have much of an idea of what features of the world can make them true or false, but the truth-values of those statements, however interpreted, will always be a strict function of the way the world is. (2) Democracy depends on the requirement that governments make sincere attempts to provide equally-treated people what they ask for. That seems to me to depend on certain political freedoms (like those of speech and assembly) being protected. But while good use of information may make for happier jurisdictions, it does not either depend on deliberative democratic schemes or make any such schemes necessary. 

Remember, people often want to effect public policies that they know full well are unlikely to make their communities happier. That doesn’t make these people undemocratic. You may support your sister-in-law as a candidate for city office because either (i) her victory will annoy a certain cousin, (ii) you don’t want her to be sad, because she’s such a nice person, or (iii) you promised her your support before you thought she had a chance of making the slate. That’s life in an actual democracy–whether a deliberative one or not. 

To conclude, democracy does not depend on the necessary existence of good people trying with all the epistemic power at their disposal to get to truths–though it may well be the case that living in a decent country requires considerably more effort along those lines than can be seen in the U.S. at present. It is, in fact, because of that sad state of affairs–in conjunction with what I take to be Lynch’s admirable attitude toward it–that I consider him such a valuable ally. Let’s face it: sharing views on democratic theory or the theory of truth is of extremely marginal importance in these dark days. What matters is retaining democracy and a love of truth–whatever may be their connections. And Prof. Lynch is absolutely right about that.

----------------

* I note that there has long been a close comradery between Hegelianism and pragmatism (especially in and around Pittsburgh) that I have long found somewhat mysterious.

# And he points out (p. 76-7) more succinctly than I have done here that Hannah Arendt’s distinction of a couple of allegedly independent types of truth is self-defeating. He does not seem to see, however, that the same sort of argument applies to his distinction between “political” and other sorts of truth. That statements of all kinds–and in different contexts–have more or fewer “political” implications does not imply that there must be various types of truth. (Although there are lots of metaphorical uses around. Consider “True dat!”)

⧫ See this paper and this blog piece for two examples of such hostility on my part .

About the Author

Walter Horn is a philosopher of politics and epistemology.

His 3:16 interview is here.

Other Hornbook of Democracy Book Reviews

His blog is here