Misled by Analysis

“Conceptual analysis” has been the dominant paradigm in English-language philosophy for much of the twentieth century. and remains influential. Traditionally, a conceptual analysis provides necessary and sufficient conditions for the proper application of a concept. For example, Porphyry defined man as a “mortal rational animal”. A little less restrictively, a modern philosopher might just attempt to lay down a few necessary conditions, or a few sufficient ones. Additionally, this conceptual analysis is supposed to be a matter of pure thought – no experiments required. In a common image: it can be done from the armchair. For a great summary and critique of conceptual analysis, see Ahlstrom-Vij’s book review of McGinn.

This has been a terrible wrong turn, in my view. Although some valuable techniques are used along the way – such as finding counterexamples to proposed generalizations – the odds of achieving true conceptual analysis are miniscule. Here I want to sketch some cognitive-psychological evidence for my skepticism, mainly so that I can refer back to it later. In other words, here I’m doing some metaphilosophy which I’ll lean on when I discuss more typical philosophical topics. A lot of my readers may find this boring, in which case by all means skip it, at least until I lean on it later and you question my reasoning. (Notice how I just implied that I have a lot of readers? Clever and subtle, huh?)

Two of the dominant cognitive-psychological models of categorization are exemplar theory and prototype theory. Take the category of “birds” for example. Both theories focus on similarity, but the exemplar theory supposes that some short list of known birds (robins, pigeons, hawks, …) provides the standard, while the prototype theory proposes that the average characteristics of all known birds provides the standard. Both theories imply that some birds (e.g. penguins) are harder to recognize as birds than others (robins). Prototype theory regards category membership not as an all-or-nothing affair, but as more of a web of interlocking categories which overlap. Exemplar theory is less committed to the existence of vague or borderline category membership, but seems at least compatible with the idea that some animals – Archaeopteryx for example – might be borderline cases of birds.

In order to be useful for thought and communication, a category need not have necessary and sufficient conditions for membership, at least not in any usable form. (A long list of features and their weightings – feathers, flight, talons, beak, etc., etc., along with numerical weightings – might work mathematically, but seems psychologically unrealistic, and certainly not discoverable from the armchair.) Instead, a workable category has short distances in similarity-space between members, compared to the distance between the category and other nearby categories. Crucially, this depends on what things happen to populate the world the speakers live in. The reptiles, mammals, and other animals we contrast to birds help make birds a useful category, because the differences between categories are relatively large. Perhaps in a galaxy far away, there is a planet with many birdlike and mammal-like animals that can be lined up in a gradually differing order with no breaks. If there are intelligent beings on that planet, it is a good bet that just one of their concepts encompasses all of those species.

Another way of saying this is: Things are clustered in the (mathematical) space of properties. The “dimensions” of this space are length, mass, number of feathers, flight speed in air, etc. – any property the thinker in question can measure or observe. And the similarity/distance metric in this space is whatever is salient to the perceiver. As members of the same species, we can usually count on each other’s similarity metrics to be largely commensurate with our own. But if we ever communicate with beings on that distant planet I speculated about in the previous paragraph – the planet of the birdmammals – we need to be more circumspect. They might perceive a glaring gap in the birdmammal spectrum that we just can’t see. Clustering is a feature of the external world X cognizer(s) interaction, not of the external world alone. External here refers to what is outside the cognizers — of course, the world as a whole includes them, which is a point philosophers could stand to remember more often.

Another reason to be skeptical about the prospects for conceptual analysis is that competence far outruns explicit reflective knowledge. We can recognize instances of a concept – pictures of cats on the internet for example – far more easily than we can lay down rules by which cats should be recognized, much less to define “cat”. The fact that it took computer programmers decades to achieve Machine Vision systems which can recognize cats (and other categories) with at least as good reliability as humans can, is evidence of how hard it is to construct such rules. And those programmers had computers to do the grunt work of logical and mathematical computation; they didn’t do it from their armchairs. And those programmers still don’t know the rules for recognizing cats – rather, they know the rules for building neural networks and the rules for training neural networks.

Two other philosophy of language related essays I really like:

Why experimental philosophy?

Experimental philosophy is an endeavor at the intersection of psychology and philosophy. In practice, it often looks like a series of surveys asking about hypothetical situations, and asking respondents to agree or disagree that in the situation, the protagonist knows a certain fact, or is conscious, or acted freely, etc. Seen uncharitably, this can look like an attempt to settle philosophical questions by popularity contest.

That’s the criticism Sean Carroll made of a brief summary by Paul Cousin of Thibaut Giraud’s work on how people think about free will.

I don’t think that’s what experimental philosophy does, at least not usually. (I speak no French so I can’t evaluate Giraud’s work.) At a minimum, experimental philosophy can act as a caution against traditional philosophical arguments which rely too glibly on the intuitions of the philosopher at some crucial point in the argument. Such experiments can support a “negative program”, in the words of Knobe and Nichols in the linked article (click on the words “Experimental philosophy” in paragraph one). By showing that the philosopher’s intuitions are not universally shared, they can raise doubts about the traditional arguments.

But I think experimental philosophy can do more than that. It can show how we got to certain “common sense” beliefs which philosophy (often with the help of science) puts into question. By carefully reconstructing the path we have taken, we can see where we went wrong, and precisely which later deductions are thrown into doubt and which are not. If scientific discoveries are involved, of course we have to understand those correctly too. Reconstructing our cognitive-developmental history is psychology. But it is also philosophy, when philosophically important ideas are at stake.

It’s easy to overlook parts of the thought-trail that got us into some “dilemma” and thus misdiagnose our problems. As I’ve argued in my previous posts on free will, free will is traditionally opposed to “determinism” because “determinism” was thought to imply universal causality, where “causality” is a one-way relationship from past events to present and future ones. But this supposed equation between determinism and universal causality is itself a scientific mistake. The traditional “problem of free will” is imaginary, and the traditional “solution” of a nonphysical mind intervening upon physics is an imaginary solution to an imaginary problem.

The “because” in my previous paragraph states a claim about the thought-trail that got us here. It is exactly the kind of claim best evaluated with a large helping of experimental philosophy.

There’s another, less important but still important use for experimental philosophy. Many philosophical problems are about how to reconcile/adapt the “manifest image” to the “scientific image”, to use Wilfrid Sellars’s phrases. In other words, supposing that we understand what science is actually telling us, how best can we state the upshots in everyday language? To know that, we have to understand how people actually use everyday language like “is conscious”, “knows”, and “acted freely”. Just asking them to set down definitions is not a good approach. (Try asking people to define “chair”, and note how few of them allow something like a beanbag chair to count.) But surveying people about hypothetical (preferably not wildly hypothetical) scenarios is a perfectly reasonable approach. If you’re interested in accurate and efficient communication – as everyone who seeks truth is – choosing the right words matters.

AI (AGI) will probably have values

Convergent evolution of the eye in vertebrates (L) and cephalopods (R). Source: Wikipedia

Some AI safety researchers (e.g. Stuart Russell) refer to an “alignment problem”. That is, they take it as highly likely that an Artificial General Intelligence (AGI) will have values, and want to ensure that those values are well aligned with human values. Here I use “AGI” to mean an AI that can do most intelligence tasks at least as well as a typical human being. In a recent Ask Me Anything podcast, Sean Carroll questioned this assumption behind the “alignment” problem: maybe AI won’t have values. There are probably many ways of achieving high performance on some tasks, he suggested; why assume that the methods implemented in AI will involve values?

I want to give two reasons why it is indeed likely that AI will have features recognizable as values, or goals. (It doesn’t matter for this purpose if what the AI has are “really” values in some deep metaphysical sense; the fact that they consistently function in a similar way is sufficient to raise and define a safety concern.) First, convergent evolution in biology has led to values at least twice. Second, the definitions of the tasks that humans will want AI to perform generally require values to understand.

The “camera” style eyes of vertebrates and cephalopods and jellyfish are a well known example of convergent evolution. The fact that multiple lineages of organisms develop the same basic solution to a problem is a good indication that the solution is a particularly good one. It would be reasonably probable to expect that were another lineage to develop independently into a niche where sight is highly useful, it too might develop a camera-style eye. This would be particularly likely if there were no known alternatives, such as compound eyes, which had also developed.

In addition to eyes, both vertebrates and cephalopods developed complex nervous systems with remarkable levels of intelligence. And both types of animals have recognizable values. They avoid danger, seek food and mate(s), and explore and play, mostly in that order of priority. Although different animals have different variations on these values – I wrote “mate(s)” for a reason – they all count as values. And there are no animals which we consider intelligent which are not guided by values in their behavior.

Maybe we’re cheating, appointing ourselves the arbiters of which animals “we consider intelligent.” But if so, we’ll also be the judges of which AI we consider intelligent, so we’re cheating fair and square.

If AGI were being developed primarily to solve, say, abstract mathematical proofs with no known applications to human life, my second argument would not get off the ground. But it’s not. AGI is being developed to assist humans with our life tasks, whether it be daily life for an elderly or injured person who needs plenty of assistance, or scientific research, or engineering, or corporate planning. Or, for a really scary thought, military strategy. But to do well on these tasks in a flexible and intelligent way, the AI has to understand what the humans want. After all it is humans, either the user or the programmer or (one hopes) both, who define what doing well on the task means. From a certain point of view – one that abstracts away from human values and just tries to describe the world “objectively” – what humans want is a very narrow and peculiar range of outcomes. And to consistently match the human-desired outcomes, the AI has to track the performance of various optional actions it could take and how well they score on these measures. There is a word for a pattern of intelligent behavior that tracks certain outcomes and makes sure that they happen. It’s called “goal” seeking. Operationally speaking, this AI will “care” about achieving these “goals.”

To repeat, for this discussion I don’t care whether the AI “really” cares, if that means for example feeling subjective emotional longing for the goal. As Edsger Dijkstra once said, “The question of whether machines can think is about as relevant as the question of whether submarines can swim.” That may not be the right attitude in all AI related areas, but when it comes to safety, it is. Whether submarines can swim or not, they can still sink your battleship.

I haven’t surveyed the reasons why AGI would not be designed with values. Maybe my readers can supply some in comments.

Who will count the votes

Я считаю, что совершенно неважно, кто и как будет в партии голосовать; но вот что чрезвычайно важно, это кто и как будет считать голоса.

I regard it as completely unimportant who in the party will vote and how, but it is extremely important who will count the votes and how.

attributed to Stalin by Boris Bazhanov, Stalin’s former personal secretary

Today (Jan 6, 2021), Congress will feature a dispute over whether to approve the Electoral College results. It is a foregone conclusion, because Democrats control the House, and it would require a majority of both the House and Senate to override the validity of the Electors that were sent by the states. Because it’s a foregone conclusion, the number of Republicans joining this putsch will be much smaller than it otherwise might have been. In other words: the symptoms will not reveal the full power of the underlying disease.

The National Archives has a useful document on the rules of the process. Part of it reads

Upon such reading of any such certificate or paper, the President of the Senate shall call for objections, if any. Every objection shall be made in writing, and shall state clearly and concisely, and without argument, the ground thereof, and shall be signed by at least one Senator and one Member of the House of Representatives before the same shall be received. When all objections so made to any vote or paper from a State shall have been received and read, the Senate shall thereupon withdraw, and such objections shall be submitted to the Senate for its decision; and the Speaker of the House of Representatives shall, in like manner, submit such objections to the House of Representatives for its decision; and no electoral vote or votes from any State which shall have been regularly given by electors whose appointment has been lawfully certified to according to section 6 of this title from which but one return has been received shall be rejected, but the two Houses concurrently may reject the vote or votes when they agree that such vote or votes have not been so regularly given by electors whose appointment has been so certified.

p. 13

So, all a faction needs to install whomever it wants as the next President and Vice President, is a majority in both Houses of Congress, and enough gall to ignore the whole democracy thing. No court would seem to have any jurisdiction over the process. All the referees have skin in the game. This seems like a glaring flaw.

Right now, we only have one party with widespread preference for their favorite conspiracy theories over the actual tallies of votes by actual voters. And even in that party, there are plenty who strongly prefer democracy. But it is not obvious why the situation in both parties will not get worse. The media are still largely following policies that encourage a race to the bottom.

After today, the pundits will congratulate us and themselves, saying the system worked. Maybe, if by “the system worked” you mean that we got lucky this time. But I can hear the ghost of Stalin (or maybe it’s just Bazhanov; all Russian ghosts sound the same to me) laughing behind our backs.

Update Jan 7: Boy, was I barking at the wrong threat! I mean, we still have to fix this Congress counts the votes thing, but only after taking better measures to stop simple thuggery.

Psychology, structures, and chemistry

Joseph E Davis has a featured post in the Aeon/Psyche newsletter titled “Let’s avoid talk of ‘chemical imbalance’: it’s people in distress.” Davis argues that “chemical imbalance” is drastically oversimplified, and distracts from more personal and more effective treatments. I think he’s basically right. (Full disclosure: my wife is a psychologist.)

How could a treatment based in verbal exchange of fuzzy human concepts and memories outperform a scientifically based treatment based on studies of the brain? Surely I’m not denying that neurotransmitters make a difference to how a person feels and behaves? Well of course not: feelings and behaviors have to be implemented somewhere, and it’s not your left pinky toe! Even if you believe in an immaterial soul that controls what you feel and do, the control has to enter the body somewhere, and the brain is the only remotely plausible candidate (if any candidate is, which is debatable).

But then, personal encounters and verbal exchanges also affect the brain. Memories are laid down by changing the neural wiring, among other possible effects. Neurotransmitters bring about signaling across synapses, but learning affects where those signals go.

Davis cites Irving Kirsch, “Placebo Effect in the Treatment of Depression and Anxiety,” whose abstract states:

analyses of the published and the unpublished clinical trial data are consistent in showing that most (if not all) of the benefits of antidepressants in the treatment of depression and anxiety are due to the placebo response, and the difference in improvement between drug and placebo is not clinically meaningful and may be due to breaking blind by both patients and clinicians. … Other treatments (e.g., psychotherapy and physical exercise) produce the same benefits as antidepressants and do so without the side effects and health risks of the active drugs. Psychotherapy and placebo treatments also show a lower relapse rate than that reported for antidepressant medication.

It’s important to remember that a placebo effect IS an effect. It can be considerably better than nothing.

If psychotherapy is so great, why doesn’t it sell better? Davis writes:

[Jenna, a depressed patient] told me she welcomed the diagnosis of a neurobiological disorder, which confirmed her problem was ‘real’ – brought on by a physiological force external to her volition – and that it showed she’s not ‘just a slacker’. At the same time, Jenna was careful to distance her experience from that of people who are, in her words, ‘crazy’ or ‘nuts’. Their illness means a loss of control and ability to function. By contrast, she sees her problem as a common and minor glitch in neurochemistry. No one, she insisted, should mistake her for the mentally ill.

The stigmatization of mental problems is the problem. Ironically, as Davis explains but I won’t quote, the “chemical imbalance” story has if anything aggravated stigmatization.

Clarke-Doane on ethics and mathematics

He’s contrasts them – kinda. So do I, but for different reasons. Here are the two bottom lines from Justin Clarke-Doane’s paper “The ethics–mathematics analogy” in Philosophy Compass 2019:

This argument is a kind of radicalization of Moore’s Open Question Argument. … The point … is that an agent may know that A is F, for any property, F, whether descriptive or ethical, while failing to endorse A. … if the argument works, it works for any normative properties, whether ethical, epistemic, prudential, or all-things-considered.

In general, if one is an ethical anti-realist on the basis of epistemological considerations, then one ought to be a mathematical anti-realist too. And, yet, ethical and mathematical realism do not stand or fall together. Ethical questions, insofar as they are practical, cannot fail to be objective in a way that mathematical questions can.

But what does he mean by “practical” near the end of the second passage? Clarke-Doane repeatedly refers to “whether to do” what the ethical (or epistemic or prudential) norm says to do. Apparently a “practical” question is one that settles whether to do X, for some particular X.

Before we evaluate whether “whether to do X” questions can “fail to be objective”, I should explain how certain mathematical questions can fail to be objective, on Clarke-Doane’s view. That is because mathematical pluralism is true of at least some mathematical domains. (I know little about philosophy of mathematics, but I must say I find mathematical pluralism highly plausible.)

Clarke-Doane: “Just as Euclidean and hyperbolic geometries are equally true, albeit true of different structures, the mathematical pluralist maintains that foundational theories, like (pure) set theories, are too. It is as though the most uncompromising mathematical relativism were true.” And: “At first approximation, mathematical pluralism says that any (first-order) consistent mathematical theory is true of the entities of which it is about.” On this basis Clarke-Doane concludes that mathematics, if pluralists are correct, is truth-bearing but not objective. I’ll take this as partially definitive of what “objective” means here. So I guess this means: if you get to pick which theory to use, it’s not “objective”.

How might one conceive or defend an ethical pluralism comparable to mathematical pluralism? Clarke-Doane asks us to consider an “ethics-like” system ethics*, which has slightly different norms and as a result tells us not to do some particular X that ethics tells us to do. Then we might wonder whether to do what ethics tells us to do in the situation, or what ethics* tells us to do. As for why ethical pluralism might be defensible, Clarke-Doane suggests that Cornell Realism implies it, as do moral functionalism and Scanlon’s metaethical views. I call my own view “Cornell Constructivism”, but that’s for another time.

Of Clarke-Doane’s two bottom lines, I agree with the first and a small part of the second. The first was that one can accept that A is F, for any normative property F, and yet not endorse it. But this undercuts Clarke-Doane’s claim in the second bottom line that ethics is “practical” in his sense. Of course it may be practical for some people – ethical people. Highly ethical people may see no daylight between concluding that an act is right, and endorsing it and going for it. On the other hand, extremely sociopathic people might see no attraction at all in the ethical. And turning to philosophical thought-experiments, it seems easy to conceive a demon who regards the ethical as a property to be avoided at any cost.

Clarke-Doane might reply that you either do X, or do not, and that is what makes it objective. But that you do X (or not) does not imply that you ever evaluated X at all. I’m really not sure what Clarke-Doane is getting at, and I worry that I’ve overlooked a better interpretation. But I can find no interpretation that truly logically connects from “ethics is practical” to “it cannot fail to be objective” and also makes both plausible.

I agree all too much that “ethical and mathematical realism do not stand or fall together” – too much to have nearly as much patience with the ethics-mathematics analogy as Clarke-Doane does. Ethics is bound up with experience in ways that make the analogy a non-starter. Ethics is about how we can flourish and get along. We who address ethical reasoning and justifications to each other. We who accept or reject these reasons and justifications, and propose alternatives. In order to determine whether our interlocutors can reasonably accept our proposals, we have to study and listen to them. In order to check whether we reasonably make the proposals, we have to study ourselves – and our common humanity will allow this to shed light on others.

Ethics isn’t a priori. It’s mired in empirical learning.

Justin Clarke-Doane has done philosophy an enormous favor by radicalizing – to the point of absurdity – Moore‘s Open Question Argument. Even Moore’s own “simple non-natural property” of goodness fails to pass Moore’s own test. We can agree that an act has Moorean Goodness and still wonder whether to do it. But if no normative property can conceivably pass the test, this shows that the test is not an appropriate test of normativity. There is no pure normativity – “pure” meaning utterly empty of descriptive content – to be had in this or any other universe.

We can endorse an action as prudent, or ethical. We can endorse an inference as logical. We can endorse a theory as epistemically virtuous. In none of these cases are we simply saying “yay, action/inference/theory!” In none of them are we purely expressing approval, or an intention to act/infer/theorize. There is additional information we are implying.

We can of course just endorse. Endorse without an “as” (as ethical, as logical, etc.). Endorsing, that is, without any value judgement. But that’s not normativity.

With apologies to Elvis

I gave my ballot to the postman / He put it in his sack

But then the man with the orange hair / Said you can’t vote while Black

He wrote upon it

Return to sender / Registration unknown

No such voter / No-vote zone

We had a pandemic / He failed at that

I write “You’re fired!” but my ballot keeps coming back

So then I took it to the post office / Machines were broken down

The postmaster general said / Can’t mail votes in this town

He wrote upon it

Return to sender / Registration unknown

No such voter / No-vote zone

This time I took the ballot myself / Down to my polling place

Where I found, to shock and dismay / Just an empty space

They wrote upon it

No people power / Democracy unknown

No such voters / No-vote zone

Southern U.S. states have closed 1,200 polling places in recent years: rights group

Postmaster General Says ‘No, I Will Not’ Put Mail Sorting Machines Back

Review: Carlo Rovelli, Reality is Not What it Seems

Carlo Rovelli is a big fan of loop quantum gravity, and of physics in general, and this book recaps the whole history of modern physics, at least partly in order to show how elegantly loop quantum gravity fits into place as a reasonable extrapolation. It’s an interesting and believable history, and the case for the plausibility of loop quantum gravity looks convincing to me. But then, I think I was an easy mark — since I already agreed with a series of strange (from the layperson’s point of view, at least) assertions Rovelli makes about known physics.

Rovelli inserts helpful diagrams every so often to summarize the history (and sometimes potential future) of “what there is” in the physical world according to physics. I can’t quite do justice to them so I use a table (please read it as one table).

NewtonSpaceTimeParticles
Faraday MaxwellSpaceTimeFieldsParticles
Einstein 1905SpacetimeFieldsParticles
Einstein 1915Covariant fieldsParticles
Quantum mech.SpacetimeQuantum fields
Quantum gravityCovariant quantum fields

In the transition from special relativity (1905) to general (1915), fields and spacetime are absorbed into “covariant fields”. This is because spacetime, Rovelli asserts (and I instinctively agree), is the gravitational field. So other fields like the electromagnetic field are covariant fields – fields that relate to each other in circumscribed ways. The curvature of spacetime depends on the energy (e.g. electromagnetic) present, and the behavior of electromagnetic fields depends on that curvature.

Rovelli likes to sum up some key features of each theory, and these summaries are very helpful. For QM, Rovelli lists three key principles:

  • Information is finite;
  • There is an elementary indeterminacy to the quantum state;
  • Reality is relational (QM describes interactions).

As a fan of Everettian QM, I don’t think we really need the indeterminacy principle. But it’s still true that we face an inevitable uncertainty every time we do a quantum experiment (it’s just that this is a kind of self-locating uncertainty).

Loop quantum gravity refines the “information is finite” principle to include spacetime as well. Not only are energy levels discrete; spacetime is also discrete. There is a smallest length and time scale. Rovelli identifies this as the Planck length (and time).

Rovelli explains loop quantum gravity as the quantization of gravity, deriving from the Wheeler-DeWitt equation. This equation can only be satisfied on closed lines aka loops. Where loops intersect, the points are called nodes, and the lines between nodes are called links. The entire network is called a graph, and also a “spin network” because the links are characterized by math familiar from the QM treatment of spin. Loop quantum gravity identifies the nodes with discrete indivisible volumes, and each link with the area of the surface dividing the two linked volumes.

Rovelli is at pains to point out that the theory really says what it’s saying. For example: “photons exist in space, whereas the quanta of gravity constitute space itself. … Quanta of space have no place to be in, because they are themselves that place.” This warning might seem too obvious to be necessary, but that’s because I didn’t reproduce the graphs of spin networks in Rovelli’s book. (I lack the artistic talent and/or internet skillz.) You know, graphs that sit there in space for you to look at.

OK, that’s space, but what about time (and aren’t these still a spacetime)? This deserves a longish excerpt:

Space as an amorphous container of things disappears from physics with quantum gravity. Things (the quanta) do not inhabit space; they dwell one over the other, and space is the fabric of their neighboring relations. As we abandon the idea of space as an inert container, similarly we must abandon the idea of time as an inert flow, along which reality unfurls.

[…] As evidenced with the Wheeler-DeWitt equation, the fundamental equations no longer contain the time variable. Time emerges, like space, from the gravitational field.

Rovelli, chapter 7

Rovelli says loop quantum gravity hews closely to QM and relativity, so I assume we get a four-dimensional spacetime which obeys the laws of general relativity at macroscopic scales.

In a section of Chapter 11 called Thermal Time, Rovelli uses thermodynamics and information theory to explain why time seems to have a preferred direction, just as “down” seems to be a preferred direction in space near a massive body. When heat flows from a hot zone into the environment, entropy increases. Since entropy reductions of any significant size are absurdly improbable, these heat flows are irreversible processes. And since basically everything in the macroscopic world (and even cellular biology) involves irreversible processes, time “flows” for us. Nevertheless, at the elementary quantum level, where entropy is undefined (or trivially defined as zero – whichever way you want to play it) time has no preferred direction. All of this will be familiar to readers of my blog who slogged through my series on free will. This is the key reason scientific determinism isn’t the scary option-stealing beast that people intuitively think it is.

There was one small section in Chap. 10 on black holes that seemed to fail as an explanation. Or maybe I’m just dense. Since spacetime is granular and there is a minimal possible size, loop quantum gravity predicts that matter inside the event horizon of a black hole must bounce. The time dilation compared to the outside universe is very long, so an observer would see no effect for a very long time, but then the black hole would “explode”. But surely “explode” is not the right word? Intuitively it would seem that any bouncing energy should emerge at a comparable rate to that at which it entered, at least for matter entering during a period of relatively stable Schwarzschild radius. Maybe by “explode” Rovelli just means the black hole would “give off substantially more energy than the usual Hawking radiation”?

The empty primary/secondary quality distinction

In a recent interview with Nigel Warburton, neuroscientist Anil Seth mentions (around 3 minutes + 30 sec) John Locke’s distinction between primary and secondary qualities. For primary qualities, the way in which it appears in our experience is pretty directly related to how things are in the world, such as solidity and movement. But there are things like colors which are secondary qualities, where the relationship between what we experience and what’s out there is more indirect and requires the participation of the observer to generate that quality.

But then, at around 6:30 in the video, Anil Seth tells us that all perception works mainly in the top-down, or inside-out direction – from high-level descriptive guesses about the world “down” to details that then fit in to or revise that picture, and from the central nervous system “out” to the periphery. From what little I know of neurology, this inside-out direction of influence is indeed quite important. But that observation threatens, or perhaps we should say trivializes, the primary/secondary distinction. (Seth may well understand this; I’m not sure. The mention of primary/secondary may only be made in order to move beyond it.)

If our perceptions of solidity and of motion are indeed primarily driven from the inside of the brain outward to the periphery, what sense can we make of the idea that our “experience is pretty directly related to how things are in the world”? Our experience is driven from guesses in the central cortical region outward, in both color and solidity experiences. It would seem that all qualities are secondary qualities.

But then, all qualities are also primary qualities, if all it takes to be a primary quality is that it can be specified without reference to an observer. For example, we can define three zones of spectral radiance, one centered at 420 nm, one at 530 nm, and one at 560 nm, each giving less weight to other wavelengths as one gets further from that peak. We can then define “red” things as those whose radiance in that highest-wavelength band bears sufficiently high ratios to the radiance in the other two bands. Of course, I had to lean on human experience of colors to get those wavelength numbers. Yet, I have to lean on human experience of solidity before I could attempt to define that, as well. The alleged primary/secondary distinction is not to be found here.

Seth points out that the solidity of a bus can impact you even when you’re not observing it. OK, but a bacterium which photosynthesizes using only rhodopsin will flourish in green light more easily than in red light of the same total intensity – regardless of whether anyone is looking. Again, no difference here.

How to start a race to the bottom

Did you hear the latest news from the courts? They’re overhauling the rules for legal arguments in front of a jury. The rules for lawyers will be much looser. Want to ask a witness an irrelevant question? One that lacks foundation? One that the witness has already answered, but you didn’t like the answer? Want to skip the questions and just testify to the courtroom on behalf of your side? Go right ahead!

The other side can object, of course, and the objections will be noted for the record. But then the questioning, or testifying, can go on as if nothing happened.

Badgering the witness? Go for it! Hearsay? No problem! Lay witness testifying about a subject he has no expertise in? Let the jury beware!

Expert witnesses also need no particular qualifications any more. If the witnesses are good enough for one side, they’re good enough for the court. It’s strictly He said, She said, from here on out. The court will not attempt to instruct the jurors regarding which witnesses are credible or have genuine expertise. Jurors will be on their own regarding whom to believe.

OK, relax. I’m just kidding. This isn’t going to happen. But if it did, it would be a disaster. Lawyers would race to the bottom to use underhanded tricks to con jurors onto their side. Truth and evidence would largely go out the window. It’s widely known that the legal rules of evidence and argument are there to prevent just such a disaster, and there is no massive wrecking ball on the horizon headed toward destroying these rules.

OK, don’t relax. Indeed, low-grade panic would be appropriate. This isn’t going to happen to the courts, but it has already happened to the press. The mainstream US print, radio, and TV media, with the exception of a few open partisans, treat “objectivity” as if it demanded a courtroom without any rules. More precisely, with only one rule: that “both sides” will get a chance to speak. And never mind how the number of sides gets magically reduced to two. Journalists have become stenographers or videographers. Fact checking is relegated to a special segment, if it exists at all. And news outlets are embarrassed if some important figures are found to be stating falsehoods on a regular basis, especially if that looks “unbalanced”.

In recent years there has been a lot of well justified hand-wringing about our post-truth society. “How did we get here?” authors ask. To me the mystery is rather: why did it take so long?

(Hat tip: Brian Leiter for the courts analogy.)