In the absence of reality-testing, bullshit wins

Too common or too costly—truth is a weak social signal

This is the thirteenth piece in Lorenzo Warby’s series of essays on the strange and disorienting times in which we live. The publication schedule for Lorenzo’s essays is available here.

Meanwhile, Arnold Kling took issue—thoughtfully, as always—with some of Essay 12 over at his place.

This one can be adumbrated thusly: Diversity-Inclusion-Equity bureaucracies measure things that don’t matter (but feel good) while ignoring things that do matter (but produce discomfort).

Housekeeping: This is the tip link; you can tip as much or as little as you like. I have complete control over tips, not substack.

Do remember this substack is free for everyone. Only contribute if you fancy. However, if you put your hands in your pocket, money goes into Lorenzo’s pocket.

This, meanwhile, is the subscription button, if that’s how you prefer to support Lorenzo’s writing:

---

In the absence of sufficiently binding reality-tests, truth is either too easy (so too common), or too hard (so too costly), a social signal.

It’s too easy when everyone—or almost everyone—can see something is true. Then there’s nothing special about affirming it.

Meanwhile, it’s too hard when it takes genuine expertise to understand why something is so.

That may give the possessors of genuine expertise both utility and status, but such expertise does not scale up socially. In the face of sufficiently binding reality-tests, you have to put the work in, otherwise the social signal will not be reliably available to you.

So, in the absence of sufficiently binding reality-tests, truth is either too cheap, or too costly, a social signal. In particular, it doesn’t differentiate, or doesn’t scale-up sufficiently, to work as a signal for “superior” in-group status.

What works much better as a social signal—in the absence of sufficiently binding reality-tests—is sophisticated falsity. Such falsity does not require the effort of genuine, reality-tested expertise but does separate some people out from the insufficiently “aware” masses. This could be because they refuse to reject the truth in front of them, or because they’re unwilling or unable to adopt rationalisations and euphemisms, or play the language games required to turn falsity into a signal for “superior” in-group membership.

Nor will any old falsity do; it has to be rhetorically convenient and congenial falsity. It has to resonate in order to replicate. It will therefore typically be bullshit: in the technical sense of statements made for persuasive effect without regard to their truth value.

Nor will any old bullshit do. It will work much better if there is some nugget of truth from which a rhetorically convenient and congenial mountain of bullshit can be constructed.

None of this works in the face of binding reality-tests. If, however, such tests are weak or absent, that means the costs of error are low. If the costs of error are low, then the efficient level of self-deception, the level of self-deception that enables us to be more persuasive and more able to moralise and rationalise our self-interest, is going to be high.

Photo by Joanna Nix-Walkup on Unsplash

This makes bullshit an even better social signal. Folk can rationalise and moralise their signalling so that congenial and convenient bullshit becomes righteous truth that all the righteous affirm. We are a very status-driven, and highly imitative, species.

So, in any social milieu with weak reality tests and lots of folk with good verbal facility and rationalisation skills—especially if there is selection-by-approval—one can expect to see the creation of mountains of moral-signalling bullshit. Which is to say, social milieus dominated by midwits. This is going to be especially so if they select for midwits.

How many of our universities, non-profits and government (or corporate) bureaucracies are precisely thus?

Do the systematic policy failures of the pandemic response begin to make more sense? Does the institutional take-up of trans madness begin to make more sense? Does the malfunctioning of our universities make more sense?

All three of these will be subjects of later essays in this Worshipping the Future series. 

The Diversity Inclusion Equity (DIE) game

For now, consider Diversity-Inclusion-Equity (usually phrased Diversity Equity Inclusion to avoid an unfortunate acronym). There is no evidence that these programs do any good. There is even some evidence that they are harmful.

Drawing attention to differences in moralising ways, and in moralising hierarchical ways, is not likely to be conducive to social harmony and human flourishing. Moreover, there is never a good reason to create inquisitors or commissars, which is what Diversity-Inclusion-Equity officers so often functionally are.

Nevertheless, Diversity-Inclusion-Equity has swept the Anglosphere and beyond. Why? Because we live in highly bureaucratised societies and Diversity-Inclusion-Equity is made for bureaucrats. It does three things bureaucracies are inclined to do.

First, it hoards authority. The bureaucracy, or key sections thereof, become fighters against bigotry, with all the moral authority that implies. Hence, trans is used by education and social work bureaucracies to undermine and supplant parental authority.

Second, it spends resources on itself. While commerce tends to select for efficiency—releasing unused or tied-up resources generates income—bureaucracy tends to select for inefficiency. The more resources spent on the bureaucracy, the better for the bureaucracy.

Third, it protects against the complexities of competence. Diversity-Inclusion-Equity justifies itself by its intentions. To criticise it is “obviously” to side with bigotry. So, no competence tests are required. It is classic “the process is the point” bureaucratic make-work. It is just made for midwits, for bureaucracies, and especially for bureaucracies that select for midwits (as they have a tendency to do).

Elite delusions

That truth is too common, or too hard, a differentiating social signal to scale up conveniently has pervasive effects in societies that are so rich and complex many people are shielded from reality tests. The further up the social order one goes, the more one can be shielded from reality tests based on the beliefs one espouses. Inevitably, the tendency is for the use of bullshit as social signalling device to increase, rather than decrease, as one climbs the greasy pole.

Davos and the World Economic Forum are, of course, the pinnacle of moral-signalling bullshit.

If one’s political benchmark is the imagined future, and given that there is no information from the future—nor, in the case of imagined futures, binding tests against self-contradiction—the effect is further magnified.

Quantifying as false signal

One common trap is that quantification can look like a reality test, yet be very much not. Instead, it’s often a mechanism for creating an illusion of selection for competence.

Consider Goodhart’s Law. The most memorable formulation of Goodhart’s Law was by anthropologist Dame Marilyn Strathern:

When a measure becomes a target, it ceases to be a good measure.

The Lucas critique in macroeconomics (if you change the conditions, behaviour is going to change) is a related idea.

If people are rewarded for achieving a particular statistical outcome, they will produce that statistical outcome. This process may extend to simple fraud, but even when statistics are honestly compiled, folk will work to score better on the statistic—as in the classic problem of “teaching to the test”.

One of the problems of universities is the use of quantified proxies for quality, such as citation rates. Academics are judged by how much they publish and are cited, not so much by how well they teach. This produces the “publish or perish” mentality, and the use of citations as a measure of success.

Citation measures have so many problems, it is hard to know where to start.

First, they do not include the most useful academic publications—good quality textbooks. Second, they encourage “salami slicing”: dividing work up into as many different individual papers as possible to increase citations, plus throwing as many darts at the dartboard as possible, hoping one will “stick”. The incentive to multiply publications also encouraged the proliferation of under-powered studies, a major cause of the replications crisis.

Third, they encourage citation multiplication. Scholars create mutual citation chains and individuals cite themselves. Both create a false appearance of acceptance.

Fourth, they enable idea laundering. A growing number of journals providing citation opportunities means that bad-but-congenial ideas can be run through the citation mill, giving the appearance of sound scholarship.

A useful dissection of problems with citations-as-target not being a good measure is here.

Quantification of this sort does not work well as a replacement for reality-tested reputation effects, which is what we evolved to use as our fundamental quality selection-and-encouragement device. Indeed, it was the systematic use of reality-tested effects to generate prestige-through-replicable-success that led to the take-off of science.

Drowning in evolutionary novelty

We live in societies of huge, and expanding, evolutionary novelty. As we are not fitness maximisers (that is what lineages do) but adaptation-executors, we are running into increasing problems with unmoored adaptations. Our evolved cognition (and metabolism) is now operating in social milieus where crucial features are primed to malfunction.

We are a highly imitative species much concerned with status. If one gets selection-by-approval, without strong character or reality tests—as is rife through our universities, non-profits and bureaucracies—then the levels of efficient self-deception will be high, as will selection for rhetorically convenient and congenial bullshit.

Meanwhile, bureaucratisation in itself weakens reality tests by insulating decision-makers from the effects of their decisions, increasing selection-by-approval and raising the levels of efficient self-deception.

Societies dominated by social signalling where truth is too weak a differentiating social signal, because it is either too common or too hard, will generate even more congenial bullshit. However, becoming ever more a civilisation of broken feedbacks is not something that can go on indefinitely, so it won’t.

If the weakness of truth as a social signal in developed societies does not terrify you, you have not thought through the implications.

The next two essays look at the consequences of the evolutionary novelty of abandoning presumptive sex roles and the accompanying feminisation of institutions and social mores.

References

Harry Frankfurt, ‘On Bullshit,’ Raritan Quarterly Review, Fall 1986, Vol.6, No.2.

Will Storr, The Status Game: On Social Position And How We Use It, HarperCollins, 2022.

John Tooby and Leda Cosmides, ‘The Psychological Foundations of Culture,’ in The Adapted Mind: Evolutionary psychology and the generation of culture, J. Barkow, L. Cosmides, J. Tooby (eds), Oxford University Press, 1992, 19-136.

Robert Trivers, The Folly of Fools: The Logic of Deceit and Self-Deception in Human Life, Basic Books, [2011], 2013.