Wednesday 27 January 2021

Aikin & Talisse on deep disagreement

A disagreement is interesting where- rather than being the predictable  consequence of heterogeneous preferences or interests- it represents a departure from the conditions for Aumann agreement- i.e. a situation where two or more parties have Bayesian rationality and share common priors and posteriors. The question arises, why do they have different types of rationality? Can they do better by changing their heuristics? Alternatively, why are their priors or posteriors different? Knowing this may help expand the common information set.

Deep disagreements are particularly interesting. It may be that they represent  phenotypic diversity or confer survival value. This points to something else Aumann has looked at- viz. why 'regret minimization' militates against unanimity and its associated 'Bayesian rationality'. The existence of Knightian Uncertainty means that a 'deep disagreement' may be associated with an ambiguity in classification of states of the world such that something 'paradigm busting' is around the corner.

It is foolish to confuse mutual hatred, or the difference between sanity and lunacy, with deep disagreement. Predictably, this is what Aikin & Talisse do. They write in 3 Quarks-
Deep disagreements are disagreements where two sides agree on so little that there are no shared resources for reasoned resolution.

This is not a disagreement. There may be no contact at all between the two sides. If there is contact, there may be mutual hatred, not disagreement.

In some cases, argument itself is impossible. The fewer shared facts or means for identifying them, the deeper the disagreement.

Only in the sense that cats and dogs have deep disagreements. 

Some hold that many disagreements are deep in this way. They contend that reasoned argument has very little role to play in discussions of the things that divide us. Call these the deep disagreement pessimists – they claim that many of the disputes we face cannot be addressed by shared reasoning.

Shared reasoning is as ineffective with people as it is with cats snarling at barking dogs. By contrast, a change in the incentive system or conditions of life may be mutually beneficial. There are ways in which you can make your cat and your dog get along such that both are happier and your home is harmonious. This happy outcome is the result of a certain type of reasoning and application of mind on your part. But it isn't a process 'shared reasoning' in which Fido & Chairman Miaow have demonstrated their ratiocinative prowess. 

It is seldom the case that matters are resolved through 'shared reasoning' because uncorrelated asymmetries exist- i.e. it is in the interest of both parties to chose different protocols or principles. But, where utility is transferable, disagreement or disagreeableness can be mitigated by some rational, economic, process or mechanism. But this is transactional, not epistemic. It is foolish to think that what makes people get along is not mutually advantageous transactions but, rather, some sophomoric debate about values or categorical imperatives. Philosophy makes a fool of itself when it pretends Society works on the same principles as an idealized academic Symposium.

There are also deep disagreement optimists. Their view is that deep disagreements are intractable only for contingent reasons – perhaps we have not yet surveyed all the available evidence, or we are waiting on new evidence, or there is some background shared methodological principle yet to be uncovered. With deep disagreement, the optimist holds, it is hasty to give up on rational exchange, because something useful is likely available, and the costs of passing such rational resolution up are too high. Better to keep the critical conversation going.

But who will pay to keep it going? Talking has an opportunity cost. The potential pay-off from talking has to exceed that cost for rational people to indulge in it. 


Disputes among pessimists and optimists regularly turn on the practical question: Are there actual deep disagreements?

No- if some state of the world, one party affirms as possible, obtains.

The debates over abortion and affirmative action were initially taken to be exemplary of disagreements that are, indeed, deep.

They are not deep at all if the Sky turns as red and blood and the Angel Gabriel appears with a flaming sword saying 'I iz gonna fuck up all youse pro-Lifers! Also, Whitey, yore ass gonna fry sho nuff!'  

Later, secularist and theists outlooks on the norms of life were taken to instantiate a divide of the requisite depth. More recently, conspiracy theories have been posed as points of view at deep odds with mainstream thought.

If Q-Anon gets hold of cctv footage showing widespread pedophilia in Federal buildings then they win. There is no deep disagreement. 


This brings us to QAnon.

Here’s QAnon’s core doctrine in a nutshell: (a) there is a cabal of Satanic and child-sex-trafficking Hollywood and Washington elites who drink ‘adrenochrome’ from tortured victims to prolong their lives; (b) Donald Trump, along with select other patriots, is waging a secret war against this group; and (c) an insider with ‘Q-level’ clearance in either intelligence or military command is leaking information out to readers on internet message boards. So Q-believers think they have insider information about a monumental war of good versus evil that’s being fought right in front of everyone’s eyes, but is nevertheless largely unseen.

Till they get hold of all that damning cctv footage which is currently safely hidden up Trump's rectum.  


Conspiracy theories typically are based in accounts of secret dealings that not only run contrary to widely accepted views but are also evidentially sealed off from them.

No. Conspiracy theories have truth makers. Sometimes, they turn out to be true in some particular and, in the face of hard evidence, the 'mainstream' has to accept that the nutjobs were right. 

Because conspiracy theories are built around a contrast between what is widely believed and what is known only by those with special access to the truth, they can thrive only among a community of conspiracy believers.

Nonsense! A conspiracy theory doesn't start off 'widely believed'. It takes time for it to disseminate. Nobody has 'special access' to the truth, though there may be specific information silos which, however, some who have access to them, believe are of great and malign significance. A UFO conspiracy theory- or a theory about the Deep State- is likely to have some proponents who actually worked in some clerical capacity within a specific information silo. Similarly, conspiracy theories about supposed medical and technological evils generally have at least one credentialized 'expert' on side. Kary Mullins is a Nobel Prize winning scientist who endorses some truly crazy shit. But he's hella smart.

Accordingly, what outsiders present as evidence against the conspiracy theory gets explained away by those on the inside.

But this is true of any hypothesis- including the one that a [articular theory is 'conspiratorial'. 

Purported evidence against the conspiracy theory is often transformed into further evidence that those on the outside are deluded, duped, and gullible.

But this is also true of the anti-conspiracy theory. In some cases, suspicions about financial institutions or intelligence agencies have turned out to be correct. There really was a conspiracy. It was the 'insiders' who were deluded, duped and gullible. 

Was the Iraq War a conspiracy to unjustly enrich certain bad actors? Or was it necessary for the War against Terror? Clearly, if there was a conspiracy, not everybody was involved. Equally, if there was no conspiracy, some bad actors did exist. Still, it may be expedient for us to decide that the thing was a Conspiracy, not a Crusade.

In the end, conspiracy theories thrive partly because in adopting them, one adopts the view that all possible evidence confirms the theory.

No. Theism, of a particular mystical sort, may thrive for this reason. But so may a sufficiently sophisticated Structural Causal Model for a narrow enough class of evidence. That is why, for the research program to remain Scientific, the scope of observation must expand. 

We can say, then, that it is part of the nature of conspiracy theories to be epistemically sealed in this way.

No we can't. The nature of Scientific or Mathematical Theories is to be epistemically sealed in a protocol bound manner. But this is also true of legal arguments or opinions given by professional experts. Without protocols, nothing is epistemically sealed. Conspiracy theories are likely to be loosey goosey. We don't expect their exponents to be sticklers for protocol.  

So it is with QAnon. Observe some popular pairings of counter-evidence and correlate sealing strategies:

Ellen DeGeneres and Oprah Winfrey have been arrested for child pornography, but they are still allowed to tape their respective shows, so as not to alert the others in the cabal.

Either this is true or it is not true. By and by, this canard won't be repeated because both women will continue their successful careers. So, there is no real 'sealing strategy'. There is just a rumor of a silly sort. 

Wayfair trunks are means for ordering and delivering abducted children to their abusers, but only select users may order them.

An urban legend is something less than a 'conspiracy theory'. It may be convenient to pretend one has fallen victim to a rumor of this sort rather than to have engaged wittingly in seditious conspiracy.  

Kamala Harris has a body double, but one cannot tell the difference on video.

This is true enough. I had to give up pant suits because I was afraid of being mistaken for her. 

Hillary Clinton was to be arrested for child sex trafficking in 2017, but she was allowed to continue her activities but under heavy surveillance.

I am not a child. Any sex trafficking which I have been subjected to since 2017 is of a reprehensible type. But it does not amount to pedophilia- though, no doubt, I wear diapers.  


These reports make no promise that there will be publicly available confirming evidence.

But, by lapse of time, they are rebutted. It's like the Mayan prophesy of the end of the World. When the world doesn't end, the thing is dead in the water. As I say, it may be convenient to pretend to believe in this sort of shit so that the quantum of punishment is reduced if your seditious conspiracy is prosecuted.  

Repeating a rumor is a type of 'cheap talk'. It is a fake type of 'common knowledge'. But it provides a sort of alibi. You can pretend you were just going with the flow- wishing to belong to a 'pooling equilibrium'. You weren't doing anything naughty or premeditated. The difficulty arises with 'costly signals' which establish a 'separating equilibrium'- in this case, one of greater culpability. An actual conspiracy may well feature a full fledged conspiracy theory. But, it would be wise to disguise it so as not to give away the conspiracy's plan of action. 

A prediction of a cataclysmic even on a given date in the near future has two effects. On the one hand it acts like a 'Kavka toxin' increasing current levels of belief. On the other hand, it will lead to 'cognitive dissonance' after the prediction fails to materialize. Some will fall away to reduce this dissonance. But those who remain will be far more deeply committed. Something like a Church is created whose members will be prepared to 'tithe' and to provide martyrs from time to time. This could be very valuable.

Aikin & Talisse don't get that a failed prediction can be a good thing. It creates a 'separating equilibrium' on the basis of cognitively costly signals. Cognitive dissonance now operates to create greater loyalty and a greater desire to win reputational advantage by a greater show of commitment to the cause. 

Thus Aikin & Talisse feel Q Anon, and- perhaps Trumpism- have been thoroughly vanquished.  


The fact that Inauguration Day has come and gone without incident is evidence that what was taken to be a deep disagreement with no shared intellectual resources for rational disputation was, in fact, a disagreement waiting for crucial testing. That test has come. Though the QAnon conspiracy theorists may elect to incorporate yet another epicycle into their worldview, that they recognize the need to regroup and revise is a minor win for reason and argument. 
It also suggests that the among Q-believers and those who reject the view is not absolutely deep after all.

The above suggests that Aikin & Talisse are as stupid as shit. There can be little 'deep disagreement' about that. Meanwhile the people who got into Q-Anon will find some other type of clickbait scam to engross their time.

1.7K
Pandemic Era Isolation and Anxiety Can Contribute to Substance Abuse

No comments: