For regular readers, a quick word of warning. If you were hoping for a blog on my usual topics of voice-hearing or trauma, you should know this has nothing to do with either. My blog may be evolving to include a semi-regular series of podcast reviews. We will see if this proves of use to me and/or of interest to anyone else…
Sam Harris and ‘failed’ conversations
Sam Harris’ attempts to have difficult conversations are an often courageous and always fascinating endeavour. They can be particularly interesting, as with most things, when they go wrong. Harris refers to these as “failed conversations”.
These typically end with Harris and his interlocutor being just as far apart as they were at the beginning of the conversation. Although a radical pessimist could see this a win, as using reason to get someone to change their views can make them hold even more extreme views, in these conversations no-one changes their mind or concedes a point, and no collaborative progress is made towards a better understanding of the world. Sometimes they just result in the interloctor defensively engaging in ad hominems.
When an episode of his Waking Up podcast results in a failed conversation, afterwards Harris will ask his audience, in genuine puzzlement, “What just happened here?”
Yet, these conversations are not failures. They help illustrate how hard the seemingly simple process of conversation can be. They help dispel the myth that you can always simply change someone’s mind by presenting them with facts and reasons. They help demonstrate strategies people use to avoid conceding they may be wrong and to avoid (publicly) changing their minds.
Sometimes simply painful to listen to (e.g., the episode entitled The Best Podcast Ever), these failed conversations, as in the case of Harris’ recent conversation with Jordan Peterson, can also be highly entertaining, albeit in a masochistic kind of way. This latter podcast caught my interest, so I’m going to quickly recap it, to try and figure out exactly what happened (to the extent time permits).
I’m not going to try and probe the philosophical underpinnings of the conversation in any depth. This has already been done, very interestingly, elsewhere. Nor will I here go into the psychology of belief change, though I may come back to that another time. Here, I’m simply going to look at the argument as it unfolded, and then give a few final thoughts.
Peterson had been doing the podcast rounds, including interviews with Joe Rogan and Duncan Trussell. When he came to speak to Sam Harris though, Peterson was always going to have a very different type of conversation.
On Rogan’s podcast, Peterson had stated that he was a scientist but also a deeply religious person. Rogan naturally asked “how do you reconcile those?” to which Peterson replied:
“Religious truth is not scientific truth. You might say, well, there’s no other truth than scientific truth… What scientific truth tells you is what things are, but genuine religious truth tells you how you should act”
This clearly conflicts with Harris’ position, as laid out, for example, in one of Harris’ TED talks, where he states:
“it’s generally understood that questions of morality — questions of good and evil and right and wrong — are questions about which science officially has no opinion. It’s thought that science can help us get what we value, but it can never tell us what we ought to value. And, consequently, most people — I think most people probably here — think that science will never answer the most important questions in human life: questions like, “What is worth living for?” “What is worth dying for?” “What constitutes a good life?”… I’m going to argue that this is an illusion… and actually quite a dangerous one at this point in human history”
The contrast between their views on this point is likely one of the main reasons so many people wanted to hear Peterson on Harris’ podcast.
What I particularly liked about the ensuing conversation, at least on the first listen, was Harris’ refusal to bow to the perceived audience pressure that he must have felt. There were a wide range of topics that could have been discussed, including morality, myth, and religion. Most hosts would have jumped over any early terminological stumbling blocks in order to get to this range of ‘hot’ topics that most of the audience presumably wanted to hear about.
But Harris didn’t do this.
He refused to move on until a basic question about the nature of truth had been dealt with.
What ensued resembled a two hour travel documentary about the USA that consisted entirely of footage of the crew at the US border being questioned about their visa. There is no way a TV producer would have ever let a documentary that turned out this way to be aired (despite the increasing plausibility of obtaining such footage), but the new media and podcasts can break such rules. If we wanted to make a parallel with the scientific process, we could see this conversation as a null finding; something that might have ended up in a file drawer. But just as we now have journals dedicated to null findings, the new media can air (seemingly) null conversations.
At multiple points during the conversation, Peterson and Harris looked like they had leapt the ‘what is truth’ barrier together, and could continue onto other things. But they kept crashing down again and again and again, as we will see.
The 1st effort: “Is it true enough?”
25 minutes into the podcast, Harris and Peterson begin to debate the question of truth.
Peterson starts by noting that, for pragmatists, if what you do works, then it is true enough. As he explains:
“The proposition that the universe is best conceptualised as subatomic particles was true enough to generate a hydrogen bomb but it wasn’t true enough to stop everyone from dying (and therefore from a Darwinian perspective it was an insufficient pragmatic proposition) and was therefore in some fundamental sense wrong.”
Harris dryly responds: “There are few issues here I think we need to pull apart”.
At this point, we can piece together Peterson’s argument, which seems to run as follows:
Premise 1: X is true if it works
Premise 2: Something doesn’t work if it leads to the death of all humans
Conclusion: X is not true if it leads to the death of all humans
Both these premises can be critiqued. Harris spends nearly all of his time addressing premise 1, but he will have a couple of quick thrusts at premise 2 as well.
Harris’ first issue with Peterson is that:
“You seem to be equivocating on the nature of truth.”
Harris reveals that he was a student of Richard Rorty at Stanford, and restates what a pragmatist would define truth as:
“Truth is whatever successfully passes muster in a conversation”.
In this sense, truth is whatever survives the selection pressures of conversation. However, Harris does not himself subscribe to this, arguing that truth does not function merely by Darwinian principles. That is to say, at the end of a conversation, conclusions that have survived the pressures of our examination and debate are not necessarily the truth. We could be wrong about the nature of reality. Indeed, Harris notes that we were not designed by evolution to be truth-establishing machines – but rather to do things such as recognize the facial expressions of apes like us.
As a result, he observes that we should expect to have our common sense intuitions “frequently and incessantly violated” by what we discover to be true about the nature of reality. What counts as a truth according to pragmatic criteria is likely to be wrong in Harris’ realist definition of truth. Harris describes how he is a scientist realist (there is a truth out there that we could be wrong about), and a moral realist (there are right and wrong answers to moral questions).
Harris is great at generating thought experiments on the spot to demonstrate logical contradictions of an argument (as we will see later) as well as going to reductio ad absurdums. It is this latter tool he employs now, in what we will call Harris Argument #1:
Harris Argument #1:
“The reduction ad absurdum of a Darwinian conception of knowledge would be if we ever learned certain truths that got us all killed, well, that would prove that these things weren’t true”
To which Peterson equivocates: “they weren’t true enough, I’d say”
Here we see the problem that will make any resolution impossible for the next hour or so. Peterson sticks to his ‘Darwinian’ pragmatic criteria of truth, whilst Harris sticks to his realist definition. Harris repeatedly shows that Peterson’s definition leads to apparently nonsensical situations, and Peterson, in response, claims that Harris’ position will also leads to problems. However, for Peterson to attempt to show this will involve a move into a discussion of morality, which Harris refuses to do until the truth issue is resolved, so Peterson never gets to make his counterpoints.
Peterson is slightly unclear as to whether, for him, a truth that isn’t true enough is necessarily false, or whether there can exist truths that are less true (compared to his pinnacle of truth which is a moral truth) but which nevertheless still meet a sufficiently high level of truthiness to be deemed true. It is somewhat confusing.
Harris is too troubled by all this to simply continue:
“This conception of truth I think we have to nail down. It just seems to me undeniable that there are facts whether or not any us… are aware of those facts… physical reality has a character whether or not there are apes around to talk about it.”
He give the example of the Trinity nuclear bomb test; before we knew about energy being trapped in an atom, it was there nevertheless.
Peterson responds: “Everything you said there I agree with”
It would hence appear the conversation can proceed. Hurrah!
But then Peterson caveats:
“My one objection to that is…is it true enough?
We felt we were in the clear, and heading to the next topic, but we have been cruelly bought back to Earth:
The 2nd effort: “I don’t think that facts are necessarily true”
To try and shed some light on Peterson’s position, Harris simply restates Harris’s Argument #1, but puts some clothes on it this time, using a scenario involving smallpox. He notes that you could accurately describe the structure of smallpox, and this knowledge could be dangerous, but:
“to point out this is dangerous… doesn’t undermine the scientific truth value”
To which Peterson replies:
“But it does undermine the claim that scientific truth is the ultimate truth”
Peterson here seems not to be saying that scientific truth isn’t true, but is downgrading it, using his continuum view of truth. But whether it remains above or below a cut-off value to be deemed truthy enough to be true, is unclear. Either way, it appears that what he is referring to is not a traditional conception of ‘truth’ at all, but something we could call ‘utility for the good life’. Harris will draw out something like this from Peterson later.
One can sense Harris’ blood pressure starting to increase as he restates Harris’s Argument #1:
“If knowing what is true got you all killed, well then, that would be a truth that wouldn’t be worth knowing, but it wouldn’t make it less true”
At this point he seems to be implicitly begging Peterson:
But Peterson does go there:
“If it doesn’t serve life, then it’s not true”
Peterson hence clarifies his position. He is no longer saying that a scientific fact about smallpox, for example, is a weaker or inferior truth, but is now saying it is not a truth at all.
It will turn out that we are not even close to getting truly bogged down yet.
And for many listeners this will turn out to be a bog they may have preferred to enter head first, to save time.
Harris pulls his arguments of the past half hour together in a sentence, dismissing first pragmatism’s and then Peterson’s Darwinian conception of truth:
“You clearly have to have a conception of facts and truth, that is possible to know, that exceeds what anyone currently knows and exceeds any concern about whether it is useful or compatible with your own survival even, to know these truths”.
To disagree with this, you are going to have to say something pretty spectacular.
Peterson doesn’t let us down:
“Ok, well then, I would say that I don’t think that facts are necessarily true”.
We have clearly entered a brave new language game.
As Harris notes in an understated response:
“You’re choosing… to use the word ‘true’ …[in a way] that will make it very difficult for people to understand what they mean“
Harris then attempts to reset things:
“I just want to make a few claims which I think are unobjectionable, and I can see whether we are on the same page here”
It will turn out that not only are they not on the same page, they are not even in the same book. And the different books they are in are written in different languages.
The 3rd effort: Yes, but…
Harris puts another set of clothes on Harris Argument #1, to see if he can get Peterson to swipe right:
“It seems to me I can make statements about reality, that neither of us can judge to be true… but we know that there is a fact of the matter. I can say, for instance, that you have an even number of hairs on your body… What do you think about that?”
Peterson says that as this could neither hurt nor kill you then he will accept this as true.
Harris is clearly setting this up to introduce some life threatening peril to hang on the ‘number of hairs’ question to see if this makes Peterson, seemingly illogically, in a realist framework, change his mind about the truthfulness of the statement about the number of hairs.
Harris does indeed do this, giving a situation in which you are taken hostage by people who have a bizarre religion in which they will kill people who have an odd numbers of hairs on their body. As he explains:
“Now your life depends on which is true about you….Nothing has changed with respect to truth-value… it’s just a different situation. Your concept of truth can’t be hostage to these superficial changes in context”
Peterson: “I think it is inevitably hostage to them.”
Harris, starting to get a bit frustrated, asks:
“Jordan, what does that mean in this context?… I think it reveals at least an awkward commitment to revising our language”
This is the essential point. Peterson is playing a different language game, and Peterson knows this. In the absence of a referee, Harris will no more be able to reconcile with him than Roger Federer can successfully play an un-referred game of tennis against a Greco-Roman wrestler who keeps catching the serve and trying to force the tennis ball into submission. No game is possible.
Peterson restates his position, which, just as Harris keeps repeating a variant of his Harris Argument #1, is a regurgitation of a point he has already made:
“I think what it means, technically is that the only final way of sorting out whether a scientific claim is sufficiently true is through Darwinian means because I think that the Darwinian process is the only way of adjudicating truth”
Harris is baffled that Peterson, a self-proclaimed scientist, really believes what he is saying:
“I’m just confused at this point… I don’t see how you can accept what you sound like you accept”
“You have to choose what you mean by truth… and I’m not accepting the same definition of truth that you operate under… I believe that pragmatism trumps realism”.
By now it is clear that Peterson does not hold his view because he does not understand it’s implications – he is holding it in spite of its implications for the present conversation.
Harris puts a third set of clothes on Harris Argument #1, and gives the example of our understanding of evolution and DNA leading us to destroy our species. This he notes, would be unfortunate, but wouldn’t imply that our understanding of evolution was wrong. Quite why Harris thinks that Peterson will be moved by this argument, which is the same in form as the previous smallpox and hair arguments that failed to change Peterson’s mind, is unclear. Perhaps it is because of the negative social implications (in most circles) of being someone who denies the truth of evolution.
Peterson sticks to his guns:
“I would say that if our knowledge of the evolutionary process was true, it wouldn’t be fatal”
Harris then tries a different argument, let’s call it Harris Argument #2. This accepts that Peterson isn’t ignorant of the point made in Harris Argument #1, so tries to make him change his mind by arguing that he is underestimating the scale of the cost of making this argument:
“I think that you are not noticing the price you are paying for redefining a word like ‘truth’…Truth is a bedrock conception… It’s that to which our very sanity is anchored and you tug at this anchor at your peril…You are marrying the concept of truth… to goodness… this creates a few problems”
Harris’ understatement is a thing of beauty.
Harris then gives the example of two groups of people who believe the same things, in one group it kills them and in the other it doesn’t. “How” he asks “could it be significant for the truth?” He goes on:
“We could put you in a situation where knowing something or not knowing something would get you killed, and yet the fact this it would get you killed doesn’t reach into the truth value of the statement”
If a potential victim also happens to give the same wrong order, and hence survives, this doesn’t make that sequence true, notes Harris.
This is a cute example, but is just a restatement of his Harris Argument #1 which hasn’t dented Peterson’s conviction in the four-or-so times he has already tried it. Sure enough, it is not effective now either.
Peterson simply responds: “It makes it true enough to survive”
Harris: “Yes, it makes it useful… it was wrong but useful”
“I’m perfectly willing to say that” responds Peterson.
Does this mean we have some agreement and we can now proceed?
Nope, our second moment of apparent escape also turns out to be illusory. This video sums up the past few minutes nicely.
The tree we run into comes from the micro/macro-examples distinction Peterson makes.
Peterson now concedes that Harris can “set up microexamples, where the claims you are making are valid”
Hmmm. I may have been premature in saying that Harris’ simple repetition with variation of Harris Argument #1 wasn’t going to help. Peterson has had to cede some ground, and has introduced a new concept to fend off Harris’ attacks; the micro-example.
Peterson then goes on to say that: “I’m not concerned with the microexamples, I’m concerned with the macroexamples”
This now gives Harris his next concept to probe…
The 4th effort: Not waving but drowning
On the heels of his success, Harris gets greedy and tries to go for a knockout, responding to Peterson by arguing:
“It’s all just micro-claims until the end of time”
“No“, says Peterson:
“It’s not all just micro-claims…It’s a massive philosophy… its not an aggregation of micro-events, although it is also that, and I would say that part of the problem with the scientific worldview… is that it doesn’t provide a reliable guide for the development of the kind of wisdom that would allow us to use our technology like grownups”
Harris then goes all Jerry Maguire on Peterson, saying:
“But” adds Harris, blocking a beautiful reunion with Peterson,”that is a change of topic… Your concept of truth is collapsing everything back to whether we survive, presumably happily”
As the last two words suggest, Harris will now change tact and go after Peterson’s second premise again. He asks why Peterson has selected survival as his ultimate good. Is this not arbitrary? There are worse things that not surviving. You can imagine situations where you could say that someone else would be better off dead, notes Harris.
At this point, many listeners will empathize with this.
Harris ends up circling back to premise 1 though:
“You have to acknowledge that something can be true and dangerous”
To which Peterson responds:
“I would say it is objectively true as far as our scientific theories are accurate at this time in this local context and that’s as far as I’m willing to go”
Is this good enough for Harris? Can we move on?
Harris: “Let’s move on with there with the proviso that we are impressively capable of misunderstanding one another.”
Yes, we’re freeeeee!
However, as our jubilation subsides and our attention returns to the podcast, we notice something has gone wrong again. They’re arguing again. How did that happen? Harris has gone back to his argument #1.
Harris: “You are committed to elevating the concept of truth… into the moral stratosphere, where it entails goodness”
Peterson: “That is precisely and exactly what I am doing.”
Harris: “But the problem with that is that then it makes it very difficult to talk about ordinary truth claims… Your conception of factual accuracy… is continually vulnerable to changes in human history that could happen in a million years. When do we get to cash this check?”.
Peterson: “I don’t know if we ever get to cash it”
We thought we were free, but at the last moment it all went wrong.
Harris’s response to this is to put yet another set of clothes on Harris Argument #1, which must be getting a bit fed up of trying all these clothes on, and probably simply wants to go home and put its feet up.
He postulates a situation in which someone writes down a 100 digit number, which ends in a 1. They then say this number is a prime number, and the largest prime number any human being has ever beheld. As Harris notes:
“That is either true or false, and it’s truth or falsity has absolutely nothing to do with the ultimate survival of the species or your personal well-being”
Peterson makes the same form of response that he has been given for the last hour, which is understandable given he has been asked to respond to the same question for the past hour:
“I don’t agree… I would say it is sufficiently true… It might be useful to talk about prime numbers, but that doesn’t necessarily mean they are true in the way that I am defining truth, which as you have already pointed out is associated with something like wisdom or goodness. It is possible it is not the sort of game that a wise person would play.”
Harris reverts to pulling on Premise 2 again:
“Then there’s the question of whether enough people survive….How many deaths would begin to erode our confidence in the ‘primeness’ of a given number? You’re forced to rewrite our entire intellectual history whether it terminates in bliss or death. It doesn’t make any sense”
He goes on:
“I need to plant a flag here. I think many people listening here…will share my frustration that you are not granting what seem to be just fairly obvious and undeniable facts, and now we are having to use this concept of truth in a pretty inconvenient way”
Peterson simply responds:
“The claim I am making is that scientific truth is nested inside moral truth. And moral truth is the final adjudicator. And your claim is, no, moral truth is nested inside scientific truth, and scientific truth is the final adjudicator…. fine, those are both coherent positions”
Harris: “But yours actually isn’t coherent.”
Peterson now tries to make his position clearer using the Irish Elk.
Here, the theory is that female Irish Elk became fixated on male antler size, meaning that, due to sexual selection, male antlers got so big that they were not commensurate with their survival. Peterson explains that:
“Then we might say, well, I guess there was something wrong with what the female elk decided to focus on, but we didn’t really know that until we went extinct. And I see that as precisely analogous to the point I am making right now. We are concentrating on certain things in a certain way, and its a scientific way… which is flawed and insufficient though very powerful, and it needs to be subordinated to something else or it will be fatal.”
Harris responds: “I can grant all of that… [but] you [Peterson] would have to say if they [facts] are dangerous to know, they’re not true… but there is so much that is inconvenient about framing it that way”
Peterson counters: “But there are equal inconveniences either way…I would say that your counter position produced just as many annoying paradoxes and complexities as my position does”
Harris: “Certainly that remains to be discovered”
Please can we go there? Please can we go onto the next topic? Please, for the love of all that is holy. Anything but another repetition of the same arguments.
Harris goes on: “This is no doubt a flaw of mine as an interlocutor but this is the kind of thing that does just drive me nuts and I want to make one more pass”
Oh sweet Jesus.
The 5th effort: Smallpox again
Harris puts the smallpox suit on Harris Argument#1 again. He gives an example in which two lab groups have exactly the same understanding of smallpox, but one group kills half the world and the other lab saves the remaining half of the world. The difference between the group is not their intentions, the reason one killed half of humanity was something trivial, a fluke, like a hole in someone’s glove. This would mean, by Peterson’s logic that one group had a true understanding of the structure of smallpox, but the other didn’t, even though they had the same understanding in mind.
Peterson tries to make it appear inevitable that the negative outcome was necessarily due to a moral lapse somewhere (neglectful suppliers, slave labour, etc). This is odd. Peterson knows that this is not going to be a productive or valid counter. Perhaps he is getting tired or frustrated. Anyone could negate this objection. Harris does so in five words:
“But it need not be”
Peterson, when pressed further about the lab members’ understanding of the physical structure of smallpox, says that:
“They were right about re-arranging the chairs on the Titanic, but they were pretty damn wrong about the fact that it might sink”
This seems to reflect Peterson’s position that the scientists had a true view of the structure of smallpox (i.e., that their micro-argument was valid), but that they were wrong about whether it was a good thing to study (their ‘higher truth’ is wrong).
Harris then tries another example to try and make the point that the fate of the species does not depend on most claims, but we still nevertheless need to be able to say that such claims are true or false:
“If someone says your wife is cheating on you… The claim whether or not she is cheating on you is an intelligible claim. That is a claim that has absolutely nothing to do with whether or not you end up killing yourself based on your reaction.”
Harris goes on to say that: “Jordan, you have to grant one thing… because you killed yourself it’s not true that she was not having an affair, that move is not open to you.”
A massive 20 second pause.
Peterson resorts to saying it “depends to a large degree precisely on what you mean by an affair.”
The Clinton strategy! Used in the context of an example involving an affair? Is this intentional humour?
He notes that “You’re assuming that the photographic evidence is prima facie evidence of an affair”
It is a classical rhetorical strategy, seizing on and critiquing practically irrelevant details of the scenario to avoid addressing the underlying point that is being made. Harris isn’t having any of it:
“Jordan, I’m just using it to demonstrate that it doesn’t make sense to subordinate our conception of truth… to what happens perhaps in some distant future vis-à-vis the survival of anyone”
Harris eventually makes the point that: “There are things that are more important than understanding reality scientifically”
Peterson: “Hey, great, that’s exactly my point”
We seem to have consensus. But, you can guess what happens next.
Harris gets back to noting that Peterson is “making it very difficult to talk about facts”
This goes on. They next dive into Hamlet but at this point I have finally lost the will to summarise. The last leap into apparent progress followed by another crash to earth has made me lose the will to continue. Stay down Steve, it’s over.
What, if anything have we learnt from this?
It is clear that Harris and Peterson share one thing; they both place great importance on truth. Harris could be described in the same way that his comrade-in-arms Richard Dawkins once was characterized, namely as having “a love of truth, a love of clarity and an almost physical discomfort at obscurity”. For Peterson truth is paramount because, as he put it on Joe Rogan’s podcast, “the alternative is hell”.
Harris, although doing a brilliant job at pulling scenarios from thin air to clarify his argument and demonstrate problems inherent in Peterson’s claims, probably did too much of this. It lead to a repetitive cycle. When it became clear that Peterson was aware of the implications of his epistemology, there was no point in simply setting up exchanges that were going to be identical to what had already come to pass. Indeed, this repetition seemed to push Peterson into using the ineffective rhetorical strategies that he employed towards the end of the podcast, perhaps due to tiredness or frustration. Peterson did, in a couple of places, concede Harris’ point (at least in the case of micro-examples). I think it was here that Harris could have made more of this point of agreement, and pushed the conversation onto the next level, without having to go for the jugular by claiming it was micro-examples all the way up.
That said, I think Peterson has to more clearly and less grudgingly acknowledge that true statements can be made about states of affairs, even if they lead to negative outcomes, and hence that truth can be divorced from goodness. Peterson should perhaps leave the word ‘truth’ alone, and introduce a new word/phrase to do the job he wants it to do, or find a word that already does this. The words ‘wise’ and ‘wisdom’ might be candidates. The process of creating the nuclear bomb clearly involved scientists finding a series of truths. Whether the whole process was wise or not (the ‘moral truth’ of the endeavor) is unclear. Perhaps whether it was wise or not will not be known until well in the future; the result is indeed hard to ‘cash in’ to use the terminology of the podcast.
However, I suspect any such move will be hard for Peterson, due to his religious commitments. In the New Testament, the term ‘truth’ (alétheia), in passages such as John 8:32 (the truth will set you free), comes with connotations of goodness, rather than just factual accuracy (indeed, in this framework Jesus is the truth; John 14:6). I imagine that once you take this conception of truth, it is hard to let go. However, I would hope that Peterson could cede enough to Harris to allow the conversation to proceed, and Harris could give Peterson enough credit to let him at least expound on this.
Anyway, I hope Harris and Peterson speak again. It may lead to another ‘failed conversation’, but it will not be a failure.