Note on Hume’s epistemology

If we leave aside his account of moral knowledge, Hume’s epistemology makes machines far better at cognition than human beings. A camera recording its inputs has impressions that never take on the faint existence of ideas, a microphone captures all sound with no filters, and various algorithms seem sufficient to relate things by resemblance and contiguity. If this is right, we can take this both as a sign of how humanizing the moral sentiments are are and how inhuman non-moral cognition is. Non-moral cognition is largely done in vain since it can never get beyond mere facts, it seeks out causes that are forever unknowable to it, and it desires a transcendental foundation it can never discover. Like Hobbes, all reasoning is mere reckoning, but unlike Hobbes there is a humanizing and refining knowledge through taste and sentiment which even discovers causal relations inside of itself.

What is most intrinsic to nature

1.) Nature = other than God.

2.) Most intrinsic = What all the actions of something converge on.

3.) What is most intrinsic to nature is impossible for it to achieve. It can be seen as able to achieve all that comes up to it, and so the most intrinsic can be viewed both as contiguous to this and as infinitely distant from it.

This is true in three senses of nature:

A.) In generation of whatever exists. Nature terminates with something common to all of nature. Absent this, there could not be one description of it. But what is common to all things cannot arise either from one of them or from the totality considered under the same description. Parmenides remains forever right that being cannot be generated. Nothing is more intrinsic to substance than its substantiality, but in converging on this they approach both the contiguous and the impossible, both the what is most immanent and what is infinitely transcendent.

This is the prophetic sense of Scripture’s depiction of nature as vanity of vanities. Vanity is innermost to nature because it cannot act except for the impossible. The subconscious world cannot even take the first step without trust in another and confidence that it will catch it as it runs headlong toward the abyss.

B.) In the generation of the human person. Though natural substances terminate their actions in substance and accident, the finite, end-terminating act of nature feeds off infinite nature: the flux and weather of chemical changes, the orbiting of the planets, the breeding and mutation of populations, the intrinsic infinity of the quantity of motion, division, extension and time etc. This dissipated infinity, which is only partially captured by the law of entropy,  can only be justified by an intrinsically concentrated infinity of spirit perfected by its proper excellence. But spirits, like existence, are not something nature can give rise to out of her own substance.

C.) The proper excellence of the human person. Human actions are made perfect by attaining a good that it cannot attain for itself, since infinite desires are satisfied only by infinite goods and infinite goods can neither be created or claimed by right. They are possessed as free gifts or not at all.

Again, morality is impossible so long as we have to chose between the intrinsic value of an action and all of the adverse consequences we can reasonably imagine arising from it. To bring these two into harmony requires that they be underwritten by the ordering action of a being that has arranged infinite accidents in advance to work out to the benefit of the person. But such a command over accidental being can only exist in the cause of being as such, spoken of in (A).

The metaphysical options

Metaphysics is discourse on being and so on what cannot not be. But then <pound the table> what about this? This is being, and it obviously can cease to exist and almost certainly will. So what about the contingent?

First, I want to lay aside a dead-end attempt to solve the problem by pointing to the difference between being and being, or being and its instantiations. When we say being cannot not be this is not because we are considering it as an abstraction. If this is all we meant, being would be no more interesting or problematic than “cat”, since taken as an abstraction and not as cat, it also doesn’t share in the contingency of a particular instance. But for being not to be is an unintelligible as a cat not to be feline, which applies as much to the concrete case as in the abstraction.  If being is anything at all even in the concrete instance it is immediately everlasting. So we are left again pounding the table and asking “So what is this then?”

Some responses:

1.) The Parmenidean. We owe the discovery that being cannot not be to Parmenides, who in turn gave one of the two simplest answers to the problem: If being cannot not be then the world given by the senses does not exist.  Sense intuition gives us only doxa or subjectively conditioned information, and discourse about the sensible never rises above opinion, i.e. an opinion that could easily be overturned by later evidence.

2.) The Kantian. Kant implicitly agrees with Parmenides that if being is true and real then the information proceeding from sense intuition cannot be, but Kant bases all knowledge on such intuitions and claims that being is a rational or logical concept as opposed to a reality that is seen in the world. If it is true that “cat” and “being” differ so radically, this is because the first is real while the other is an abstraction from the real.

While Kant was not an analytic philosopher his account of being is the inspiration for the dismissive or “thin” accounts of being that characterize that tradition (with the notable exceptions of BV, Milton Munitz and Barry Miller). The trivialization of existence that one finds in the Analytic tradition – the shepherd going out in the morning to count his non-existent sheep or existence as little more than a comment about the number of things one has – is the mirror image of the Parmenidean trivialization of natural science. Metaphysics is not about anything, but only a gesture in the direction of things we should get to know by physical science and (maybe) moral theory.

3.) Triangulation theories. The Parmenidean and Kantian-Analytical traditions mark out the two extremes of the question of being by making it either real to the exclusion of the contingent-sensible or a purely logical predicate. Most theories try to triangulate these two extremes by carving out some sort of reality for being and the contingent.

a.) Participation theories.  In one sense every triangulation theory is a participation theory. More narrowly, we’re talking about the Platonic theory that sees contingent reality as an accident of an eternal form projected onto a receptacle. As an accident, it has the power to remind us of a the form that we have already known in some prior way, but the things of the physical world are not substantial. Aristotle’s linguistic account of substance as what is not predicated of or present in another does not target any ontological reality but is only a linguistic description of things that are insubstantial in themselves.

b.) Complex-substance theories. Here having a form is seen as making things substantial, though they are still derivative entities to things that are purely formal. The substantial entities depend on the purely formal ones in either the order of agent or final causality, and therefore as upon a purely extrinsic cause. While these theories preserve our intuitive sense of the substantiality of the world, they make it dependent on purely formal being only in the order of generation. Once a substance is generated, it is as substantial as purely formal being.

c.) Creation theories. The things of the world are substantial, but this substance has an intrinsic dependence on an extrinsic, purely formal reality. This occurs because this purely formal reality is responsible not only for the generation of the substance but for its act of existence. All natural generation therefore converges on a formality that it is incapable of achieving of itself, even though it is the fulfillment of the very process of generation.


Source of diversity in Metaphysics

To see the intelligible nature of anything is to know what cannot be otherwise. This usually leaves contingency intact: a gerbil can’t be a non-gerbil but it can cease to be altogether. But it seems that metaphysics is an extended discourse on the exception to this since being cannot not be. The varied responses to this are the deepest fault lines in metaphysics.

Well, if that shows it isn’t magic…

Any sufficiently new advanced technology will be indistinguishable from life (cf. mechanical shows and moving temple-statues, clocks, steam engines, recording tape, microchips… We just have to make them complex enough and BANG!)

Nagel ad 1

A: Broadly Darwinian and Materialistic accounts of consciousness, cognition, and morality are unconvincing, sometimes to the point of incoherence.

B: But look at all the success the Materialist assumption has had! It is simply crazy to turn our backs on it before we have given it more of a chance. Your argument is frustrating and uninteresting.

A: Yes, but a Post-Materialist theory would have by definition even more success and evidence in its favor. Shouldn’t you feel like you’re impeding progress by telling us not to look for a more advanced theory?

Short theology

A possible being is one that needs something other than itself to exist. Since all parts fit this description, necessity is proportionate to simplicity. This is why necessity is a term of analysis, which strives to get to that which is just itself and nothing more.

Material analysis can never give us something that is self-identical. It is not even a matter of the whole being greater than its part,  but the whole being other than its part. Simplicity is known to us only in the analysis to self, and so the self is the paradigmatic simple reality. Necessity and simplicity are therefore both proportionate to each other, and both to the reflective existence of the self.

Power is to be a source of another and so it is inverse to possibility and thus proportionate to necessity. Thus necessity, power, self-reflection and simplicity are all proportionate.

Glory and majesty in turn arise from a self with power..




Facts as copies

Leibniz: A fact (or a contingency that happens to be actual) is a sort of copy, and so has a double relation to both a previous copy and an the original.

Communicated act is the source of fixity in the contingent and of knowledge.

Leaving aside errors, infinite copies of an image are possible. If God had his interns copying forever it might even be the case that every copy has a previous copy. But this does not do away with the original, which is thus both at an infinite distance from what we now have and immediately present to it.

If God is the original of a line of copies then by definition he contains no errors since these are nothing but deviations from the image. God is immediately all perfect, though in a way that has nothing to do with the “moral perfection” that we hear about God having.

Errors become possible because the relation of God to the universe-image is not the relation of the Father to his Image.

It’s easy to be impressed by how Leibniz is ahead of his time – it’s hard to explain what he is speaking of without talking about subconscious knowledge, programs reading information, information machines, etc. But he seems ahead of our time too by anticipating what it will take to have a true universe of copied information.

God is a source of information, but information is formal. In the world of “it from bit” things no longer can act on each other. Yet things clearly reference each other by their actions so as to form a unified whole, and so God becomes necessary as a primal programmer.

Leibniz seems to back off the idea of all truths beings pre-programmed into the monad, yet a purely formal system doesn’t seem to allow for this. Matter might do the trick.


Divine benevolence

-Divine benevolence makes its first appearance in modern philosophy with Descartes saying that this benevolence makes it impossible that God give him non-veridical cognitive powers.

-Descartes, in this as so much else, sets the tone for what will count as divine benevolence, i.e. creating humans with veridical cognitive powers. As an account of divine benevolence, this is ridiculously specific and even hubristic. God could not have possibly had another more important overriding good than giving us veridical sense powers? Really? What if God thought it was more perfect to have species, and all their attendant powers, arise by selection and chance? What if he thought the freedom of superior intellects who were able to deceive us was just as important as our own freedom?

-There are completely convincing proofs of divine benevolence, but all of them prove that God perfectly acts in a way appropriate to a god.

Ah, but there is one morality that is common to all intellectual agents, and God is an intellectual agent! This makes less sense than insisting that there is one morality for all sentient agents, and concluding that dogs are therefore evil for not marrying (like men) or not eating their spouses (like spiders).

Ah, but we know that God could never torture a small child without a good reason. If he did this, it would defeat any ability to call him “benevolent”! Here again, we’re appealing to this one-morality-for-all-intellectual-agents idea. Taken more closely, the question (which was Paul Draper’s) has a torture clause and the “without a good reason” clause. Remove the second, and the first becomes (at best) far less convincing, but the second says nothing more than that a thing with intelligence acts with intelligence. Again, all we get is that God does whatever his intelligence discerns is proper for gods to do. This certainly makes him benevolent, but it tells us nothing about how he acts towards us.

-God’s challenge to Job can be re-imaged as the command: “Go ahead and tell me about this ‘divine morality’ you seem to so confidently know about. Tell me about all the observations of the divine life that you’ve based this on.” You might as well try to predict the behavior of any species that you’ve never observed.

-Descartes should have concluded that reason is incapable of establishing an absolute basis for itself and, given it desires one, can only borrow it from the divine mind. He could have made an impressive defense of philosophia ancilla theologiae. If you want to know anything about God’s plan for you, or for humanity you have no means to get it unless he tells you. Reason can’t see the end of history.

Reasoning about the unseen

You could probably get Hume to admit to these Leibnizian claims:

Sufficient reasons are evident from terms.

Facts are never evident from terms.

Analyses of facts that are bound to what is sensible yield only facts.

Leibniz  adds the PSR to these and gets a cosmological argument, since to break out of the bounds of the sensible is to posit the trans-cosmic. Hume’s whole epistemology requires that analysis be bound to the sensible, which requires some denial of the PSR.

There is probably less distance between these two opinions than at first blush. Hume’s discussion of fundamental causes is not characterized by rejection and denial but by a sort of humility in the face of what can’t be known and a corresponding refusal to talk about it. The springs of natural action are simply unknown to us and should be respected as unknown. But note the first thing Leibniz says about the PSR:

32. [by the PSR] there can be no fact real or existing, no statement true, unless there be a sufficient reason, why it should be so and not otherwise, although these reasons usually cannot be known by us.

Although these reasons usually can’t be known by us. The cosmological argument is an inference from a world largely unknown and perhaps unknowable, and so with certain facts that we can analyse no further. It’s not as if Leibniz’s vision of our epistemological predicament is one where the human mind has cast infinite light and intelligibility in all directions, and then at the far end of a set of Euclidean interences one trumphantly concludes to God and his attributes. The inference to divinity is one that we make from a world that is mostly in the dark, perhaps permanently so. Theism is not a seeing into nature but a judgment about a mostly unseen world.


« Older entries Newer entries »