Featured image is Sunset, by Caspar David Friedrich.

This post is dedicated to Drew Summitt, who has relentlessly pushed Aristotelian metaphysics upon me. It is also a technical followup to this piece.

To have beliefs, one must have a lot of other beliefs. This is John Searle’s summary of the point that, in analytic philosophy anyway, goes back at least as far as W. V. Quine. No lone belief is coherent in isolation, but only as part of a web of beliefs that provide it with context.

Rather than a web, Searle spoke of a Network. At first he believed the Network was a set of unconscious beliefs that provide context for conscious beliefs. But in time he came to see that the notion of an “unconscious belief” is dubious. Instead, we ought to speak of having the capacity to generate some specific belief.

We think of memory as a storehouse of propositions and images, as a kind of big library or filing cabinet of representations. But we should think of memory rather as a mechanism for generating current performance, including conscious thoughts and actions, based on past experience.

(…)

Instead of saying “To have a belief, one has to have a lot of other beliefs,” one should say “To have a conscious thought, one has to have the capacity to generate a lot of other conscious thoughts. And these conscious thoughts all require further capacities for their application.”

The Network is the specific set of capacities for generating the relevant beliefs. It is a subset of the Background, which are all of the non-mental capabilities that generate mental states.

I find this taxonomy compelling. I would summarize the basic insight as follows: consciousness, knowledge, beliefs, and all mental states are performed, not stored. As Richard Moodey put it, “I imagine ‘knowledge’ as inseparable from acts of knowing, as something performed, rather than possessed.”

So we have performed mental states, and we have capacities for generating them. What is the ontology of these capacities?

A Directed Metaphysics

I would suggest that an Aristotelian framework, drawing on potentiality and actuality, as well as efficient and final causes, is best suited to explain this.

Potentiality and actuality is one of those relationships that seems very intuitive at first blush but very quickly gets dizzyingly complex when you dig into the details and begin to see the number of distinctions that must be made. My guide in this is Edward Feser, a modern apologist for the classical (that is, scholastic) system. One simple distinction we can make here, following him, is between different levels of each. Consider someone who is fluent in English and someone who isn’t, say a Spanish speaker. For the Spanish speaker, the potentiality to speak English is of a higher level than the fluent English speaker’s is. That is, it is the potentiality to learn human languages in general, something which the English speaker has as well, but he also has the lower level potentiality that comes with fluency in a specific language (just as the Spanish speaker has for Spanish).

More to the point, consider a Christian and an atheist. The potentiality for the atheist to have the conscious state of believing in the resurrection of Christ is of a higher level than the potentiality for a Christian to do so. Just as the non-English speaker must become fluent in English to move to the lower level potentiality, so too must the atheist be persuaded to be, well, not an atheist any more.

Efficient causes will be the most familiar concept in this scheme to the modern reader. They are what initiates a change. Thus adding water to a sponge is the efficient cause of it becoming engorged. But in the classical system, efficient causes are unintelligible without final causes. Final causes are the directed aspect of causation. Feser puts it like this:

Indeed, for the Scholastics, even the simplest causal regularity in the order of efficient causes presupposes final causality. If some cause A regularly generates some effect or range of effects B—–rather than C, D, or no effect at all–—then that can only be because A of its nature is “directed at” or “points to” the generation of B specifically as its inherent end or goal. To oversimplify somewhat, we might say that if A is an efficient cause of B, then B is the final cause of A. If we deny this—–in particular, if we deny that a thing by virtue of its nature or essence has causal powers that are directed toward certain specific outcomes as to an end or goal—then (the Scholastic holds) efficient causality becomes unintelligible. Causes and effects become inherently “loose and separate,” and there is no reason in principle why any cause might not be followed by any effect whatsoever or none at all.

Thus it would be more accurate to say that a change is caused by the connection between what precipitates the change and the end state of the change. And the end state of the change is an actuality that was only possible because of a prior potentiality.

Imagine a scenario in which the smell of a flower reminds you of a specific time in your life. You had the potentiality to remember that time before smelling the flower. The smell is the efficient cause, and the actualized memory is the final cause.

Descartes and other early moderns banished finals causes, as well as actuality and potentiality, from our metaphysical discourse almost entirely. Leibniz preserved the latter somewhat through the modern notions of potential and kinetic energy.

I think that these concepts from the classical system are at least defensible. And I think that Searle’s philosophy of mind provides a good application for them. The Background is a system of pontentialities for particular conscious states, the Network is the specific subset of potentialities for intentional states. Each of the conscious states is a final cause for a range of efficient causes which can actualize them.

Communion as the Locus of Knowledge

There is one more facet of all of this I would like to explore before wrapping up, and that is the social dimension.

Moodey’s critique of Harry Collins provides a good entry point for this. Collins, responding to the individualist denial of the existence of social entities, argued that “the primary subject of knowledge is the group, not the individual”. Moodey replies that only an individual can be a “performer of knowledge”. Outside of the performance, knowledge is not something that any subject possesses.

However, he concedes that collectivities are the locus of knowledge.

So I can agree with Collins that the locus of knowledge is always a collectivity, sometimes as small as a collectivity of two persons. What I mean by this is that persons perform acts of knowing in social settings. Even if the person who performs the act is physically separate from other persons, others are virtually present.

The lower level potentialities for conscious states are shaped by the social games that we play. Consider Charles Taylor on the way that infants learn to speak a language:

But the fundamental point that emerges from the ontogenesis of language is that it can only be imparted from within relations of shared emotional bonding, what we might call “communion”. Language cannot be generated from within; it can only come to the child from her milieu— although once it is mastered, innovation becomes possible. The young child grasps a word that is proffered to her from the parent. She has to catch on to and follow the communicative intent of the adult.

The game, or communion, between infant and parent has an asymmetry of footings, but this is to the infant’s benefit—they can slowly draw on the richer vocabulary and discernments of the adult. The games played by adults usually require more of them, but of course asymmetric, or at least variable, footings and perspectives remain. Just as the Network of beliefs and other intentional states is made possible at all through the games an infant plays with adults (thus allowing the development of language capabilities), so too do the games we play throughout our lives shape our Networks along the way.

That is how I read Moodey’s remark that, even when other people are not physically present, they are “virtually present” to every performance of knowledge. Knowledge is always performed individually, but in the context of social games.