Informal Omega Inconsistency

Sad Carnap -- ``I guess it just doesn't
feel great that other thinkers got real
pictures and actual discussion of their
work whereas I am reduced to a literal
actual cartoon in this blog post?''
Just a quick blog post writing up a thing that no doubt people in the informal logic literature know a lot about but which I don't know the term for so I invented my own. I call it Informal Omega Inconsistency, in honour of Rudolf Carnap's logical work. I find it... the fallacy not Carnap's work... very annoying. As I shall argue below, I think we should all be prepared to see a lot of Informal Omega Inconsistency after the US election.

Informal Omega Inconsistency is when people agree to a general (existential) claim but will stubbornly deny or remain absurdly sceptical as to every particular instance of it you produce. So, somebody may well agree that there are bad drivers in Pennsylvania -- but every time one points to a particularly erratic person on the road in the state they will say that, no no, this is not a bad driver, this is somebody whose car has suddenly and inexplicably stopped working, or is cursed, or at least they will not believe it is a bad driver till these possibilities have been ruled out, or... whatever. Just for some reason every instance that might witness the existential claim granted turns out not to be granted as an actual instance, no matter what lengths must be gone to deny as much.

Sounds wacky, right? Maybe, but I think it will be easily recognised as a very common by anybody who has ever argued about racism. Of course everybody will agree there are racists, certainly, it's still a terrible problem and there are lots of liberal pieties I could complete this list with that would gain equally near universal assent in my social circles. But this or that particular instance? Oh no, you have to understand, he's a very kind soul, you must be misinterpreting what he meant by ``All coloureds must die'' -- maybe he was talking about a novel method of rendering crayons reusable? And, look, he really likes dress up even months after halloween, so that was probably just a ghost costume, and of course he's a very devout man so he likes to build crosses wherever he goes, but alas he's a smoker (nobody's perfect!) so he probably was getting his lighter out then he tripped and fell and it just happened to set the cross ablaze, and....

I parody, but not by as much as you'd like. Lots of people are Informally Omega Inconsistent and it's super annoying. I think what prevents more general recognition of this fallacy is two things. First, it's a fallacy that is only recognisable in aggregate. On any one occasion it's consistent to deny that this witnesses one's general claim -- it only becomes Informal Omega Inconsitency once it's apparent that this is a matter of policy, that this is how the person always responds to apparent instances of the general claim being made. Second, for reasons that are a bit opaque to me, we tend to think that people `want' to make the strongest claim they can, so it seems that if somebody wanted to make the general claim they'd be only too happy to grant some instances -- but not so, as this experience has taught me.

In the other logical direction, so to speak, we can also get fallacious reasoning. This is where somebody affirms a universal generalisation but comes up with some ad hoc excuse to explain away any particular apparent counter example. This is the well known No True Scotsman fallacy. I suspect that this fallacy is better known because in some sense the logical error is immediate in one case -- if you affirm a universal generalisation then deny an instance you are there and then contradicting yourself. The Informal Omega Inconsistent reasoner, on the other hand, has on no particular occasion shown their hand.

Xunzi -- ``But why though?"'
Anyway here is why I think we are going to need this concept around when we are done with this election. A lot of people have said that one thing this election has done is made it no longer possible to reaasonably deny that America is a place riddled with various prejudices. Quite so, I expect the general claims that such prejudices are rife shall henceforth receive more ready assent from a wide variety of the population. But my overall prediction is that what we shall ultimately see is more Informal Omega Inconsistency, rather than more productive dialogue -- in just the same way that Victorian sexual mores lead to more hypocrisy, rather than less shagging. As I reflect on this election, and everything about it, and everything I expect to result, I am basically just all the more convinced of Xunzi's great maxim: human nature is bad.


  1. Great phrase! I will help it spread!

  2. Well, I think it does matter what penalty you are trying to levy onto them. "There's a lot of people around who are kind of racist" is a phrase that'll find broad agreement, but if you add "and they all should be fired from their job and subjected to mandatory eight hours a day sensitivity retraining" (first example that came to mind because you said racists, not *racism*, implying that you think the problem is that some people are just bad apples), anyway suddenly people will end up discovering that oh, these racists are probably much more rare than you think because they do not have a clear division between factual and moral claims and if you make a moral claim they disagree with, then trying to drive down the factual claim that is plugged into it is a completely acceptable strategy. Which may even be pragmatically sensible! So maybe don't ask questions in ways that makes people pragmatically wary of you if you want unbiased, factual answers.

    1. I'm fine with this, since I think it follows from considerations about inductive risk - - that you should change how much evidence you require before you accept a claim depending on how serious you take the consequences of error to be in the area. However I would *urge* people who do this to base their idea of what the consequences are on solid evidence about specific consequences actually at stake in their context. *Not* generic statements about the overall state of the culture and political correctness and what not, *not* YouTube videos with cherry picked examples of SJs Gone Mad by The Panzer Skeptic Who Is Rational (or etc), and *not* just generic fears and insecurities one always has about social faux pas. Actual evidence of what would actually happen here, in this case, if we came to believe that something was racist - this is what should guide our sense of the appropriate evidential standards.

  3. I read this when you first wrote it and have just come back to it after seeing friends do exactly what you describe. One in particular--an otherwise reasonable person--denied that there's evidence that Blake Neff is racist.

    I have a further hypothesis. Trumpism makes people omega-inconsistent in other domains. There's actually a kind of left-wing Trump support that has manifested itself through a degree of this kind of inconsistency in commentators like Glenn Greenwald, Matt Taibbi, and Michael Tracey since around 2018. "Yes, of course we condemn Trump. But we don't condemn his handling of the coronavirus pandemic. Or his fascism/authoritarianism. Or..."

    In both kinds of cases (racism and Trump support), you might be able to find one instance of the general claim that the person will recognize. So what we're talking about is a degree concept, of which formal omega-inconsistency is an idealization.

    I'm sure there are other domains where we see this.

  4. Commenting 5 years after you posted this, but I just googled "omega inconsistency", and this was one of the top results!

    Interestingly, I don't think that informal omega inconsistency is always irrational. I will happily concede that there are some truths I don't believe, but I can't produce any witnesses to that existential claim. If I was in a position to identify an unbelieved truth, I'd start believing it. I'll also deny any witnesses you propose, either because I believe it, or because I don't think it's true.

    It feels like your cases are different, of course. I think the relevant difference is this: in your cases, if the existential claim were true, you'd expect there to be some identifiable witnesses. So the people who endorse the existential claim should also expect to find themselves finding positive witnesses. When they don't find themselves doing that, that's evidence that either the existential claim is false or else they're not appropriately categorizing the particular cases.


Post a Comment

Popular posts from this blog

How I Am A Marxist

Facts vs Opinions

The Flavour of Truth