In Defense of Motivated Belief

In Defense of Motivated Belief

You open Facebook. It’s 2012, and at the top of your newsfeed is another post from your uncle, an image, telling you that Barack HUSSEIN Obama is a Muslim from Kenya secretly trying to impose Sharia law on the US. You’re not quite sure how this squares with yesterday’s post that Michelle Obama is transgender biological male, or last week’s post about Obama’s commitment to the Gay Agenda, but you wonder whether your uncle really believes any of this. Of course you can’t ask him directly – he’ll insist that he does until you’re both blue in the face, as you found out last Thanksgiving – but you suspect that the factual truth of these memes doesn’t rank highly on the list of reasons for sharing them.

You open Twitter. It’s May 2020, and the same 200k-follower bluecheck who was telling you in March that it’s racist to worry about the coronavirus (because worrying about the coronavirus is something right-wing silicon valley nerds do), and in April that we should definitely be very worried but that masks are counterproductive and don’t even work (because stockpiling masks is something crazy right-wing preppers do), is now dunking on a 300-follower troglodyte who’s annoyed at having to wear masks in the grocery store (because not wearing a mask is something only a selfish right-winger with no sense of the common good would do). Each successive shift has been seamless and full of sanctimony, without even an acknowledgement of a shift at all. You begin to wonder if actual facts might not have much to do with the constant appeals to “scientific consensus” at all, but if it might have more to do with signaling – “hey, I’m one of the good guys.” (and you quickly suppress the half-formed thought that maybe actual good guys would care about actual facts, since that’s something a right-wing contrarian would say)

That’s what I’m going to be defending here. No, not either of these two particular instances, but the more basic tendency toward motivated belief which in these two cases has been wrested from its proper and healthy context with results ranging from annoying to disastrous. Addressing these kinds of failure modes will require – not the intellectual asceticism and the commitment to literal meaning of the Rationalist faith – but a reestablishment of that proper context, within which we can engage in motivated belief with eyes open.

The Coup of Literalness

Today, raised in a scholastic-industrial culture, the idea that the point of language is to convey literally and verifiably true statements about the world is second nature to us. Language can of course be used for other things, but these we call “literary”: perhaps using fiction or poetry or metaphor to convey “deeper truths” or some such platitude, but clearly separated in principle from the more basic literal use of language. From this perspective, the opening two examples – the phatic use of language; making factual assertions not because they’re literally true, but in order to signal group loyalty – would be an abberation; a sin, even. Or as Harry Frankfurt famously called it, BS.

And so, the bluecheck calls your uncle an unsophisticated anti-intellectual rube. And your uncle calls the bluecheck a corrupt and venal striver. And it’s hard to disagree with either. Hypocritical though we may be in practice, we all agree on the standard of literal truth in principle, at least for the purposes of dunking on the outgroup.

But language was not understood, or used, in a primarily literal way until relatively recently in human history. It is not that the literary dominated the literal, but that no practical distinction was made until the rise of the W.E.I.R.D. mind and its methodical, analytical approach that we recognize today as scientific rationality. When a Rudolph Carnap dismisses theology and metaphysics as “mere language games” – and even when more moderate antipositivists assert that statements do not have to be falsifiable to be meaningful – both of these are assertions about the function of language quite at odds with any premodern use of language, beyond particularly literate urbanites. Incanctations with no obvious physical referents, metaphysical shibboleths, are no abberations; they were for most of human history the primary function of language.

So how did humans ever interact successfully with the world if they were relatively unconcerned with the literal truth of linguistic statements?

In fact, this unconcern did hinder a great deal of human planning, as can be seen by the low and more-or-less-constant level of economic development in every human society prior to the Industrial Revolution, when W.E.I.R.D. consciousness first came into its own. And of course the unconcern was not absolute. In post-neolithic societies, ultimately, the crops had to grow.

But on the whole, the human evolutionary niche was one of exploiting the environment through collective action. And in this endeavor, preventing free-riding was usually the more stringent constraint than the environment itself. Language had to coordinate the growing of the crops, yes, but more importantly it had to bind together a community with common norms and who could trust each other; to instill noninstrumental loyalty to a normative community with whom one undertakes collective action. And because it’s never possible to motivate everyone in a large enough group to cooperate simply on the basis of their own objective interests, language could not be purely literal. Under these circumstances, the tradeoff of literalness – giving up the power of ritual and incantation, in exchange for a more effective and systematic approach to provision – was not worth it.

Thus, the fact that language can be used to convey factual information must be understood as a side-effect of a faculty for establishing common normative understandings. To understand language as being for the conveyance of factual information makes its evolution an insoluble puzzle. When would it ever be worthwhile to take a costly action based on information given to you by someone else, given that they could be lying? And so the evolution of irrational trust – the social impulse, and the establishment of a normative community through shared ritual to undertake collective action – must have preceded the use of language to convey factual information.

In the modern era, the dominance of literal language could only have come about, first, in a situation where trust could largely be taken for granted (high trust is a feature of W.E.I.R.D.ness), and second, where the collective action aspect of life has receded. This latter has been made possible by high-capacity states and an extensive money economy, where economic needs are provided not through personalistic collective action, but through impersonal dyadic exchange. The more the collective action aspect of economic provision recedes, the less the danger of free-riding, the more worthwhile it becomes to trade off collective action potential for a systematic epistemology oriented to literal, material truth: hence the scientific and industrial revolutions.

Is Phatic Language Even Valuable Anymore?

So far we have an explanation, but not a defense, of motivated belief. A rationalist might even read this story and conclude even more strongly in favor of literal language. After all, if we have high-capacity states and an extensive money economy now, if the “freeing” of language to convey literal meaning has unleashed so much economic and intellectual potential, surely it would be a mistake to go back on that tradeoff? If our economic provision no longer depends on it, surely the remnants of phatic belief – the foibles of the uncle and the bluecheck – should be discouraged as much as possible?

It’s true that, now that phatic and factual language can be clearly distinguished, representing the former as the latter – as in the first two examples – is exasperating, even dishonest. But two points suggest there’s a baby concealed in the bathwater we’re so eager to throw out.

First, the necessity for collective action has not been eliminated by high-capacity states and extensive money economies. The burden of collective action has simply fallen onto specialists. Those specialists in policymaking and administration cannot get by without the sort of solidarity that results from phatic language, which is why it is seen so often in politics. One would hope that that solidarity is to a national ideal rather than to partisan interests, and it may manifest in more or less healthy ways, but it must exist all the same. The luxury that the layperson now has to dispense with solidarity, itself depends on the solidarity of these specialists in upholding that liberal order.

More important for the layperson, however, is the fact that humans are obligate community-formers, much like they are obligate language-acquirers. Even if the objective material need to form personalistic normative communities and engage in collective action has waned, the subjective need has not. Like a cat raised without certain visual stimuli, like Genie who made it through childhood without linguistic stimulus, the social stimulus of a normative community engaged in collective action – and in particular, an experience of intense socialization early in life – is necessary for healthy development and continued normal functioning.1

Unfortunately, the distinction between phatic and factual language is a sort of loss of innocence. To understand phatic speech as phatic speech is to understand it as not factual. The literalist, broadly speaking (who believes literally) and the rationalist, broadly speaking (who does not) are agreed on this point. And however frequent the failures and peccadilloes in practice, as in the opening examples, most moderns are fully bought into the distinction in principle and thus either rationalists or literalists.

Despite this agreement, the rationalist and the fundamentalist cannot form a community together if literal belief is understood to be the prerequisite. But the rationalist can form explicitly value-based associations. PTA groups, book clubs, hobbyist groups, even activist groups. Thus the question: can stable, healthy, and fulfilling normative communities be constructed around value statements specifically understood as not factual? Is ‘I value’ enough, or must such a community also cohere around ‘I believe’?

Value-centered communities, for whatever reason, seem to face an iron tradeoff between intensity and scale. People can be very intensely devoted to parochial interests, as any participant in local politics will know. At the smallest scale, friendship has the potential to be intense – and it is, to be sure, important to healthy socialization – but it is concrete and inward-focused, not really an example of collective action (indeed, the strongest friendships tend to form inside normative communities). On the other end of the scale, sending $10 to the NRA or the ACLU signals allegiance to some cause and grants membership into a large community, but it is an anonymous community with no power (or, at best, an ersatz power) of socialization.

The function of ‘I believe’ rather than ‘I value’ is to overcome this tradeoff. Note that activist groups make precisely this jump: the milquetoast ‘I value the environment’ of the Sierra Club versus the apocalyptic ‘I believe we have five years before climate change makes the earth uninhabitable’ of Extinction Rebellion. ‘I value racial equality’ of Obama-era civil rights versus ‘IN THIS HOUSE WE BELIEVE‘ of the post-Trump reaction.

If this is true, it presents the rationalist – as I presume most readers are, in a broad sense – with a dilemma. If intense socialization is a necessary formative experience, if ongoing normative community is necessary to live a happy and fulfilling life, and if the strongest normative communities are oriented around statements of belief rather than of values, what’s a rationalist to do?

The Possibility of Instrumental Community

Ultimately the normative imperative associated with community-formation is simply: do. And apparently ‘we believe’ is a much more potent inducement to doing than ‘we value’. In the past, however – before the reification of reasons – community has been organized around a simple ‘do’, with neither an explicit ‘we believe’ nor a ‘we value’. Families, to some extent, are still like this – even in the Western world – and indeed values and beliefs are ways of getting people to behave as if in a kin group, with non-kin.

Of course, the modern values reasons, and will not generally undertake deliberate action without them. The modern feels at home in a value-based community, if he shares its values, or a belief-based community, if he accepts its beliefs. Simply ‘do’ is not enough.

And indeed, a community formed around ‘do’ will be underdetermined and unstable. In a subsistence environment, ‘do’ entailed survival. Hunting, agriculture, and so on: the ‘do’ was provided by the environment and required no particular justification. But in the modern environment, where survival is more or less a matter of course, just what do we do? Reasons and motivation are crucial – at least at the community level.

But for an individual, it is enough to accept one’s own human nature and the need for normative community. Even though contrary to one’s ‘objective’ interests in terms of (say) money, time, or reproductive fitness, deliberately committing to a normative community will lead to a subjectively healthier state of being. Whether or not that type of subjectivity is stable on an evolutionary timescale, each of us existing humans has a set of developmental imperatives. Or as Aristotle would say, one’s nature implies a telos. To fulfill one’s own nature, a community must be joined, and its purposes committed to, for no other reason than that they are its purposes.

An entire community likely cannot subsist this way, at least explicitly, without either disintegrating or falling prey to more virulent purposes. But an individual rationalist – resolved to accept the necessary irrationality for the sake of his own soul – can perhaps be saved this way.

To do so successfully requires a disciplined reorientation; a willingness to understand phatic statements as such. To participate in a ritual, to recite the Nicene Creed (“We believe in one God, the Father, the Almighty, Maker of heaven and earth…”) – these must be understood not as factual assertions at all which might be literally true or false, but rather as declarations of one’s membership in a normative community, and an expression of solidarity in its purposes.

This feels inauthentic to the modern; even deceptive, given that such statements have the form of factual language. And in a literalist community, where factual belief is central – sola fide – indeed it would be. But not all communities are so. In fact, a preoccupation with authenticity is downstream of the same modernist distinction between the factual and the phatic. Authentic speech accurately represents one’s internalized factual and normative commitments; inauthentic speech does not. But in a community where the doing is enough – one of orthopraxy rather than orthodoxy – to participate for its own sake is neither authentic nor inauthentic. The distinction simply does not apply. (And indeed, authenticity is a neurosis of modernity that we would likely be better off without anyway)

Submission with Eyes Open

Having been weaned on Sola Fide, the primacy of conscience, and the priesthood of all believers, I recall being puzzled in discussions of institutional churches. From a literalist perspective, and considering a creed only factually, how could one call something ‘belief’ at all if it was offloaded onto an authoritative institution? How could the doing be enough – doesn’t it matter why? To submit one’s judgment on these matters to a community, or to an institution, would seem like an illegitimate abdication.

But if such statements are understood neither authentically or cynically, but phatically, doctrinal submission is no intellectual compromise or abdication of conscience. Extra ecclesiam nulla salus – “outside the Church there is no salvation” – can be made sense of this way: not as a tautology (where believing is what makes you part of the true Church), and not literally (where nonmembers go to Hell – so unfair, ugh!), but that the creed anchors the bonds of community in a way that doesn’t make sense outside the context of that community. To be “spiritual but not religious” is to eat the colorful plastic packaging while discarding the meat inside.

In fact, because factual belief is held as an individual whereas phatic belief is held as a group, both rationalism and literalism make community otiose on its own terms. The community might persist by accident, but a community formed on sola fide – the primacy of factual commitment – cannot admit its own necessity except incidentally. Thus, literalism tends to degenerate into “spiritual but not religious” solipsism over time, and rationalism – despite the fact that its presuppositions are shared with varying degrees of rigor by most of the modern world – hardly constitutes a normative community at all outside the Bay Area and a handful of Yudkowski acolytes.

But just like a community needs concrete direction on what to do, it will not do to just join any community. Submission cannot be offered haphazardly, and phatic speech undisciplined by a healthy and personalistic normative community has dire failure modes, as in the opening examples.

First, while phatic submission may be to an organization of any size, it must be practiced in a personalistic and not an anonymous community. A church, a commune, a kibbutz – all of these may be large organizations of varying formality, but in all of them, membership entails actively participating in collective action with a community. Importantly, this responsibility cannot be commuted financially without destroying the personalistic quality that makes such a community valuable in the first place. The NRA and the ACLU, the Republican and the Democratic parties, national identity in general – these are gravity wells, attracting and dissipating the sort of loyalty that ought to be directed into a personalistic community, providing some addictive sense of belonging or purpose in the short run, but ultimately an empty one without personal socialization. The endeavors toward which they are oriented may be substantively valuable once the need for personalistic community has already been satisfied. But as substitutes for personalistic community, when the distinction between factual and phatic language blurs on their behalf, they tend to overgrow their legitimate boundaries in the mind and consume it. This is the uncle’s pitfall.

Second, personalistic communities may be more or less healthy. Intense socialization can be an important formative experience, especially during the teenage years, but one too-easy way to distill intensity is to center on one cardinal value. This can be harder to see if it is couched in a belief statement. But without an ecology of irreducible values (again, perhaps couched in belief statements), normative communities are liable to spiral into a piety contest dynamic, where “competing statements are judged, not on the basis of their accuracy or coherence, but by the degree to which they reflect some sacred value.” It is no wonder that activist communities, cults, and in general communities that have recently sprung from an anomic wasteland, tend to be the most “toxic” in a variety of ways, as opposed to older communities in which the monstrous implications of any single value have been weathered down by other values (though it is worth noting that older communities can also become toxic if rationalization or anomie sweep away these bulwarks).

Third, in a scientifically rationalized world, phatic commitments must be carefully separated from empirical questions. To take a stand on faith against the validity of evolution, or of the effectiveness of masks, or the outcome of genetic research, is to abdicate one’s conscience illegitimately, and it is this danger that the proper community of rationalists is rightly so vigilant against. Roy Rappaport notes that sacred propositions most usually “have no material referents, they are not amenable to verification, but neither are they vulnerable to falsification,” so this is not a modernist fencing of the sacred. It must be remembered what kind of institution is competent to make what kind of pronouncement, and using what process. This is the bluecheck’s pitfall.

Communities with any of these features should be consciously avoided. Phatic submission, therefore, is not unconditional submission. If it is not the individual’s place to evaluate the community’s phatic commitments one by one, it is the individual’s place to evaluate on an ongoing basis whether those commitments as a whole can anchor a community to provide the desired fulfillment, or whether it is likely to – intentionally or not – parasitize one’s mental energy.


There remain many places where the distinction between factual and phatic remains resolutely unclear. Like the opening examples, many of these are instances of annoyance. But as the basis for a eudaimonistic approach to life under liberalism, they do offer a glimmer of hope that, through discipline, the W.E.I.R.D.est of us can avoid both the temptation to create a world inhospitable to everyone else, and the temptation to lapse into anomie ourselves.

In particular, the modern preoccupation with reasons and authenticity must be held loosely, particularly where phatic belief is most valuable. The unbelievers in fact may yet be redeemed back into communion, and the true believers may yet recognize the first-order value of healthy community, as well as the preconditions of it.

Credo in unum Deum, Patrem omnipotentem, factorem caeli et terrae, visibilium omnium et invisibilium.


  1. Consider the systemic consequences of the fact that, for most children in the developed world, the public school system stands in for a real normative community.


Civil SocietyReligionRoy Rappaport


Facebook Twitter Reddit StumbleUpon

Leave a Reply

More Content