The Clustering Problem

How and why do properties cluster together into an object? How and why do many accidents join together to form a single substance? How do they constitute a cohesive unitary whole? The properties are separate existences yet they form clusters: what binds them together? What is the unifying glue? Why don’t they occur independently and singly, and why don’t they fall apart for lack of necessary connection? It is different for analytic connections: there is no puzzle about how and why bachelorhood and maleness and singleness cling together, but why do shape and color (say) keep each other company? What makes an object combine being red with being square?[1] How do these different things belong to the same thing? Parts of objects can exist separately, but properties of objects can’t: no property can be instantiated without other properties being instantiated alongside it. Properties necessarily come in pluralities; a certain sort of holism obtains. In fact, a given object has a huge number of properties if we include relations; it is a hive of properties. What makes them band together so tenaciously? There is no force that we know of that does this—nothing like magnetic attraction or superglue. They hold tightly together but without visible support. Yet they don’t merge or meld. They come in neat bundles but there is no mechanism of bundling. It seems quasi-miraculous—they just do it! In a world without clustering, it would seem like a miracle if someone brought it about—like bringing all the stars together into a single enormous super-star. How does Plato account for it, with his universals all lined up discretely in a row? Do they suddenly embrace each other? Predicates don’t do that: you can write one predicate down without having to write down any others. Even concepts are not so group-oriented: you can think one concept of an object without having to think other concepts of that object (except analytically connected ones). Whence this propensity to coalesce that we see in properties? They seem to seek out each other’s company, as if they can’t stand to be alone. What is this curious connective affinity? What is the solution to the binding problem?

Someone might feel so baffled by this binding that they contemplate denying the appearances. It isn’t really true that a single object has many mysteriously connected properties. There is no monism of objects at the center of a plurality of properties; rather, there is a corresponding plurality of objects. There isn’t one object that is both red and square; there is one red object and one square object. Each object has its characteristic property, but no object has both. It is like the mind and the body for a Cartesian: the mind is thought and the body is extension—nothing is both.[2] Similarly, nothing is both colored and shaped; different things are. We go dualist on objects, thus avoiding the clustering problem. There is no such thing as clustering, only a misguided monistic ontology. Away with the multiply instantiating object! We expand our ontology to save our metaphysics. Or again, we wax eliminative: the only true properties are shape properties—colors are mythical, non-existent. Then there is no clustering to worry about (compare eliminative materialism). And there is a third option: we go full monism–all properties are identical! We don’t have to worry about the mind-body problem (that mysterious psychophysical nexus) if all properties are physical; we just reduce the mind to the body. Similarly, we might claim that all properties are of a single type when you get right down to it—as it might be, shape properties (modes of extension). Colors are really shapes (of molecules, say). Then there is no clustering of disparate properties: everything can be done with macro shapes and micro shapes, with no one thing having more than a single property. Obviously, this is a pretty dramatic move, but it exists in logical space (I don’t believe it for a second). Or we could get even more dramatic and deny that anything exists—no objects and no properties. The whole ontology of substances and attributes is misguided, an illusion of reason. Then there will be no clustering of properties in objects, just the appearance of it. If the clustering is mysterious and inexplicable, as we naively think of it, then philosophers of a certain stripe will seek for extravagant solutions, generally revisionary (mystery phobia can lead to strange abreactions).

The problem of causation is that we can’t identify and describe the necessary connection in which we think causation consists. The problem of analyticity is that we can’t discern the semantic entailment on which it rests. The mind-body problem is that we can’t understand the emergence relation between brain and mind, though it apparently exists. The problem of clustering is that we can’t grasp the principle of cohesion that ties one property to other properties in a single object. We have a family of problems here of similar form, all centering on opaque necessary connections. My purpose has been to add the clustering problem to the list. This is a problem in basic ontology—it could hardly get more basic. It is the problem of how objects are possible; alternatively, what properties are.[3]

[1] Geometry treats shape independently of color, thus showing their independence, but shape cannot exist concretely without color—or mass, position, solidity, rigidity, etc.

[2] It is odd that the mind-body problem has not generated more than two substances (matter and spirit), given that the mind is not a homogeneous domain. Why not postulate one substance for reason and another for sensation, given the deep differences between them? Or one for volition and one for cognition. How could such different attributes coexist in a single object? Yet they appear to.

[3] Hume called causation the cement of the universe—what holds the whole contraption together. But there is a more fundamental kind of cement—the kind that holds individual objects together. Properties don’t just co-exist in a single object; they cannot be instantiated without the assistance of other properties. Why is this and how is it accomplished? That is the problem of objects themselves—how the bundling of properties comes about, by necessity. They are not really bundles (loose assemblages) but more like living organisms—complex organized internally cooperative things. It is as if the properties are in a symbiotic pact with each other. Colors and shapes are born cooperatively into an object in which they live out their days till separated or the object is destroyed. In some ways they are like cooperating genes with objects as their vehicle. They stick together opportunistically. They cohabit in a single object for the duration; they have no choice.

Al Franken and Me

I was at George Soros’s wedding in 2013, a lavish affair with hundreds of people in attendance, talking to Senator Al Franken. I explained George’s favorite joke to Al: What is the difference between a Hungarian and a Rumanian? They will both sell you their mother, but the Hungarian will deliver. No bad, Al remarked. This was before Senator Franken had been removed from the Democratic party; he was still a promising new senator representing Minnesota. I myself was in the process of being cancelled by the American philosophy profession. A couple of years later his own party destroyed his political career in their ill-advised liberal zeal. At this point I lost faith in the Democratic party, as I had already lost faith in the American philosophy profession (remember I was not the only philosopher to lose his job and reputation). This was American liberals at their worst. I don’t doubt that this kind of behavior—stupid and cruel—is what led to the troubles of the Democrats, which led to the triumph of Trump. I also think that the treatment of various philosophers around this time is a disgrace to American philosophy that will not soon be erased, if ever. It’s the kind of thing that makes people despise universities. We know what happened next: Trump and the rise of the punitive right. It is notable that Al Franken is still rejected by the Democratic party and the cancelled philosophers are still cancelled. I see a bleak future ahead in politics and university life. The politics of personal destruction is alive and well on the left.

Extremism and Violence

People (pundits, politicians) have been trying to figure out the origins of political violence. We are told that such violence stems not from the radical left or the radical right but from being too radical. We need more political moderation, less “radicalization”. The dictionary (OED) gives us the following for “radical”: “relating to or affecting the fundamental nature of something—innovative or progressive”. It is hard to see what could be wrong with that: surely, we can be concerned about the fundamental nature of something and be innovative and progressive in regard to it. Reading on we find as the third definition: “advocating thorough political and social reform; politically extreme”. The first part is again unexceptionable: sometimes thorough reform is desirable and welcomed (e.g., various forms of tyranny). It is the second part that people have in mind in speaking of being overly radical: we mustn’t be extreme. Extremism is the problem, the cause of violence. We must educate people to be less extreme in their opinions (perhaps also in their emotions)—more restrained, qualified, compromising. But is that right? Isn’t extremism sometimes warranted? Wasn’t Jesus pretty extreme? If something is very wrong, you need to be extreme about it—you can’t be mealy-mouthed, wavering, compromising. I myself am quite extreme about genocide: I think people who do it are behaving very badly (and are probably bad people—I would not hang out with such people). I am an anti-genocide extremist. Aren’t you? In fact, I am an extremist about many things: torture, animal cruelty, rape, murder, etc. I think these things are really bad and I am completely against them, extremely so. There is nothing wrong with extremism as such, obviously. It all depends on what you are extremist about. Racial discrimination is clearly bad, or the death penalty for naughty children, or a boss who works his employees to death—we should be extremely against such things. It is right to be an extremist in support of the good, but not in support of the bad. In other words, you can’t keep substantive moral values out of the definition of what should be discouraged. Obviously! There is no morally neutral definition of what kind of extremism should be tolerated. It isn’t how strongly you believe, it’s what you believe in. If you believe in evil things, you are apt to do evil things—and they are likely to be violent acts. If you falsely believe that all people of a certain ethnic group are murderers and rapists, you are likely to do harm to members of that group—you might even advocate deportation for anyone of that group. If you have false ignorant beliefs about the badness of innocent people, you will be apt to do violent things to them. Not so if your beliefs are true and rational. So, we need not teach moral and political moderation; we can teach instead moral and political rationality. Accordingly, if one side of a political divide has many irrational false beliefs, while the other side has mostly rational true beliefs, then the latter should be applauded and the former criticized. The right policy, then, is simply this: instill true beliefs and discourage false beliefs. Then you won’t get people acting on false moral and political beliefs, sometimes violently. We can’t avoid the hard work of truth and logical reasoning by simply banning any form of “extremism”. For sometimes extremism is good and anything short of extremism less than good, even bad (“Oh, I’m not so opposed to genocide in all cases, though I agree it can be quite bad sometimes”).

But this is not the end of the story. It isn’t just a matter of beliefs; we need to reckon with emotions, traits of character, streaks of insanity, criminal tendences. Some people are more prone to violence than others, with beliefs held constant—they just tend to hit out more. The roots of political violence, then, are not confined to moral and political beliefs, but include other psychological factors—and these need to be controlled too. What are these violence-inducing psychological dispositions? That is a hard question, but it is not to be answered by talk of being “radical” or “extremist”. These mantras are far too simple, papering over a complex psychological reality. But better to acknowledge this than offer threadbare formulas like “Don’t be a radical” or “Don’t go to extremes”. This is political punditry not genuine intellectual engagement with the issue.[1]

[1] It is sometimes said that the internet is the problem, but it can only be part of the problem. I read things on the internet but I don’t get hijacked by crazy or evil ideologies; the reason, presumably, is that I am too well-educated to be so easily fooled. So, better education must be part of the solution.