Facts and Incomplete Information

A modern notion of hard-nosed common sense is to appeal exclusively to positive facts. This is also a major basis of simplistic notions of empirical science. Serious scientific methodologies are more indirect, and quite a bit more involved.

From a broadly Kantian point of view that I think Plato and Aristotle would also endorse, all putative facts are really just assertions of facts, made by people. The validity of the corresponding assertions depends on the soundness of the reasons that lead us to believe them. Thus, regardless of whether our concern is ethical or scientific, it is always the quality of reasons that matters in assessing the validity of assertions.

The notion of a fair and objective weighing of evidence for or against an assertion presupposes that we symmetrically consider the pro and con, as Plato emphasized in his discussions of “dialectic”. But the simplistic bias for positive facts results in an inherently asymmetrical treatment any time we have to deal with incomplete information, because what putative facts we currently have in our possession is in part a matter of sheer chance.

In a fact-biased approach, if there happen to be insufficient facts in our possession to adequately support a hypothesis, the hypothesis is likely to be be dismissed out of hand as “speculation”, regardless of how inherently plausible it might otherwise be. We end up assuming something is not true merely because we cannot empirically prove it is true. This is independent of any other prejudice that may also enter into situations involving human judgment.

Once again, I want to recommend a prudent suspension or qualification of belief in cases of incomplete information, rather than active disbelief. (See also Debate on Prehistory.)

Husserl on Perception

“External perception is a constant pretension to accomplish something that, by its very nature, it is not in a position to accomplish. Thus, it harbors an essential contradiction, as it were. My meaning will soon become clear to you once you intuitively grasp how the objective sense exhibits itself as a unity in the unending manifolds of possible appearances; and seen upon closer inspection, how the continual synthesis, as a unity of coinciding, allows the same sense to appear, and how a consciousness of ever new possibilities of appearance constantly persists over against the factual, limited courses of appearance, transcending them.”

“Let us begin by noting that the aspect, the perspectival adumbration through which every spatial object invariably appears, only manifests the spatial object from one side. No matter how completely we may perceive a thing, it is never given in perception with the characteristics that qualify it and make it up as a sensible thing from all sides at once. We cannot avoid speaking of such and such sides of the object that are actually perceived. Every aspect, every continuity of single adumbrations, regardless how far this continuity may extend, offers us only sides. And to our mind this is not just a statement of fact: it is inconceivable that external perception would exhaust the sensible-material content of its perceived object; it is inconceivable that a perceptual object could be given in the entirety of its sensibly intuitive features, literally, from all sides at once in a self-contained perception” (Husserl, Analyses Concerning Passive and Active Synthesis, pp. 39-40).

Adumbration is something like foreshadowing.

While many of his contemporaries were caught up in the logical empiricist enthusiasm for literal “sense data” as the supposedly rock-solid foundation for knowledge, Husserl was taking an extremely original approach to a more classical view of the inherent limiting and “transcending” features of sense perception, explicitly bringing out implicit characteristics of any possible seeing of physical objects that seem clear as soon as we bring them into focus and reflect on them.

We need not take something like Plato’s refusal to treat sensation as a source of knowledge as a case of repugnance toward physicality. With Husserl’s help we can “see” a more specific grounding of Plato’s view in reasons inherent to the subject matter. Husserl’s exceptionally clear examples in the realm of visual perception also provide a kind of model for understanding something like Hegel’s frequent complaints against “one-sided” points of view.

“When we view the table, we view it from some particular side…. Yet the table has still other sides” (p. 40). “It is clear that a non-intuitive pointing beyond or indicating is what characterizes the side actually seen as a mere side” (p. 41). “In every moment of perceiving, the perceived is what it is in its mode of appearance [as] a system of referential implications…. And it calls out to us, as it were, in these referential implications: ‘There is still more to see here, turn me so you can see all my sides, let your gaze run through me, draw closer to me, divide me up; keep on looking at me over again and again…'” (ibid).

“These indications are at the same time tendencies that push us toward the appearances not given…. They are pointers into an emptiness since the non-actualized appearances are neither consciously intended nor presentified. In other words, everything that genuinely appears is an appearing thing only by virtue of being intertwined and permeated with an intentional empty horizon, that is, by virtue of being surrounded by a halo of emptiness with respect to appearance. It is an emptiness that is not a nothingness, but an emptiness to be filled-out; it is a determinable indeterminacy” (p. 42).

“In spite of its emptiness, the sense of this halo of consciousness is a prefiguring that prescribes a rule for the transition to new actualizing appearances…. This holds time and again for every perceptual phase of the streaming process of perceiving…. There is a constant process of anticipation, of preunderstanding” (pp. 42-43).

“[A]s soon as a new side becomes visible, a side that has just been visible disappears from sight….But what has become non-visible is not cognitively lost for us…. Having already once seen the back side of an unfamiliar object and, turning back to perceive the front side, the empty premonition of the back side now has a determinate prefiguring that it did not have previously” (pp. 45-46).

“The fact that a re-perception, a renewed perception of the same thing, is possible for transcendence characterizes the fundamental trait of transcendent perception, alone through which an abiding world is there for us, a reality that can be pregiven for us and can be freely at our disposal” (p. 47).

Here “transcendence” just refers to the various characteristics of the incomplete perception of spatial objects he is pointing out.

“[W]e see that every perception [implicitly] invokes an entire perceptual system; every appearance that arises in it implies an entire system of appearances” (p. 48). “What is already given to consciousness in a primordial-impressional manner points to new modes of appearance through its halo which, when occurring, emerge as partly confirming, partly determining more closely…. Advancing along this line, the empty intentions are transformed respectively into expectations” (p. 49).

Perception gives us the very opposite of isolated sense data. Every perception is connected to other perceptions.

“If we ask, finally, what gives unity within every temporal point of the momentary appearance… we will also come across reciprocal intentions that are fulfilled simultaneously and reciprocally” (p. 50).

Substance in the elementary sense of something persisting through change emerges from networks of mutually reinforcing cross-references.

“We can never think the given object without empty horizons in any phase of perception and, what amounts to the same thing, without apperceptive adumbration. With adumbration there is simultaneously a pointing beyond what is exhibiting itself in a genuine sense. Genuine exhibition is itself, again, not a pure and simple possession on the model of immanence with its esse = percipi [to be = to be perceived]; instead, it is a partially fulfilled intention that contains unfulfilled indications that point beyond” (p. 56).

“[I]n the process of perceiving, the sense itself is continually cultivated so in steady transformation, constantly leaving open the possibility of new transformations” (p. 57).

Everything we perceive reaches beyond itself, raising new questions.

“We always have the external object in the flesh (we see it, grasp, seize it), and yet it is always at an infinite distance mentally. What we do grasp of it pretends to be its essence; and it is it too, but it remains so only in an incomplete approximation, an approximation that grasps something of it, but in doing so also constantly grasps into emptiness that cries out for fulfillment” (pp. 58-59).

I suggested above that what Husserl illustrates so clearly about visual perception can serve as a model for other things. In particular, I think both facts and beliefs share the perspectival character of visual perception of spatial objects, because they revolve around analogous issues of correspondence with something external.

The very best and most complete facts about anything at best resemble a collection of still views of a tree from different angles, like the sides of the table in Husserl’s example. The virtue of facts is that they are supposed to be individually self-contained, and individually verifiable by correspondence to states of affairs. Even leaving aside all questions of interpretation that tend to unravel this putative self-containedness, by virtue of their isolation all individual facts still remain “one-sided” or perspectival, like individual still views of the tree.

Even the most complete collection or sequence of still views fails to capture the simultaneous many-sided unity-in-diversity of the concrete tree. The real concrete unity of the tree is not factual but teleological and “transcendental”, forever out of reach of a merely factual approach.

If this is true of the best possible facts, I would say it must also be true of the best possible beliefs, because both revolve around a kind of correspondence to states of affairs. The difference is that beliefs are just assertions of correspondence between what we say and what “is”. But to qualify as a fact, an assertion must also be verifiable by correspondence.

But verification by correspondence can only apply to what appears, not to what “is”, so facts only apply to what appears about states of affairs. Facts in effect just are verifiable appearances. They are an instance of what Plato called “true opinion”. They are objects of justified true belief, and potentially of a kind of subjective “certainty”.

Beliefs on the other hand usually reach beyond appearances toward what is, so although they assert a kind of correspondence, they cannot in general be verified by correspondence. Their well-foundedness in the general case has to do with a goodness of reasons. Well-foundedness by reasons falls short of certainty in one way, but it reaches deeper. It is potentially less subject to perturbation, because it does not directly depend on appearances or correspondence.

I think knowledge is something stronger than well-founded belief. Unlike facts and beliefs, I want to say that knowledge in the proper sense has nothing to do with correspondence to something outside itself. Also, well-founded beliefs may depend on assumptions that could eventually be refuted, but “knowledge” in the sense I want to give it does not depend on any assumptions either.

Contrary to common usage, then, I want to say that facts are not knowledge, and even certainty about appearances is not knowledge.

Judgments of correspondence — including beliefs and facts and certainties about appearance — seem to me to be inherently perspectival in the way that Husserl talks about. On the other hand, that rare thing called knowledge, in the way I am using the term, would be immune to perspectival limitations, because it does not depend on correspondence at all. (See also Husserl on Passive Synthesis; Opinion, Belief, Knowledge?; Sense Certainty?; Taking “Things” as True; Berkeley on Perception; Platonic Truth; Everyday Belief; A Criterion for Knowledge?; McDowell on the Space of Reasons; The Non-Primacy of Perception.)

Heidegger, Sartre, Aquinas?

The heyday of existential Thomism is well past, but Etienne Gilson and others were certainly not wrong to take note of a close connection, despite other large differences.

Heidegger in Being and Time (1926) famously claimed that philosophers since Plato had been preoccupied with questions about beings and had lost sight of the central importance of Being writ large. Many 20th century Thomists partially accepted this argument, but contended that Aquinas was an obvious exception, citing Aquinas’ identification of God with pure Being. Heidegger rejected that identification, and would have insisted that Being was not a being at all, not even the unique one in which essence and existence were identified. Nonetheless there is a broad parallel, to the extent that Heidegger and Aquinas each in their own way stress the dependency of beings on Being.

In some circles, Aquinas has been criticized for promoting a “philosophers’ God”. But according to Burrell, Aquinas argued in effect that on the assumption that there is only one God, the God of Summa Theologica and the God of common doctrine must be acknowledged to have the same referent even if they have different senses, like Frege’s example of the morning star and the evening star.

Sartre in his 1945 lecture “Existentialism is a Humanism” put forth the formula that “existence precedes essence”. Aquinas in Being and Essence had argued that God has no essence other than existence. Sartre argued in effect that the human has no essence other than existence. In his context, this is to say either that the human essence consists only in matters of fact, or that there is simply no such thing as a human essence.

Sartre’s use of the word “essence” reflects a straw-man caricature of bad essentialism. Whatever we may say that essence really is, contrary to Sartre’s usage it is supposed to be distinguished from simple matters of fact. On the other hand, in formal logic, existence does reduce to matters of fact.

What Aquinas, Heidegger, and Sartre have in common is that they all want to treat existence as something that transcends the merely factual and formal-logical. Speaking schematically, it is rather the analogues of essence that transcend the merely factual in the Platonic and Aristotelian traditions. Thus Aquinas made a major innovation in inventing a new, unprecedented concept of existence that transcends the factual. I’m inclined, however, to sympathize with Dietrich of Freiberg’s argument that the concept of essence could already do all the work that Aquinas’ new supercharged concept of existence was supposed to do.

What is important for practical purposes is that there is something that transcends the merely factual. I think the close connection of “essence” with form and ends makes it an ideal candidate. The big difference between form and ends on the one hand and facts on the other is that logically speaking, facts can be arbitrary, whereas any form or end or essence necessarily implies some nonarbitrary order.

For Aquinas, God is simultaneously a fact and more than fact, and is unique in this regard. Nothing else has this dual status. Sartre transferred this unique dual status to the human. By contrast, the neoplatonic One is strictly more than fact — in traditional language, the One as source of being was said to be “beyond being” altogether. The 20th century theologian Paul Tillich quipped that it could be considered blasphemy to say that God exists (because “existence” is mundane and factual).

The “To-Be itself” of Aquinas, while profoundly innovative with respect to previous tradition and certainly not strictly Aristotelian, is nonetheless arguably more Aristotelian in spirit than the neoplatonic One, insofar as it is less ambiguous about the goodness of the actual world. Plotinus struggled mightily to reconcile a commitment to the goodness and beauty of this actual world with an ascetic tendency to devalue all finite things in face of the infinite One. In Aquinas there is still some tension between the reality of secondary causes and the absolute dependence of everything on God, but I think it is fair to say that the way Aquinas sets up the problem makes the reconciliation easier to achieve. This was a huge accomplishment. Nonetheless, taking into account other factors like assertions about the place of omnipotence and sheer power in the scheme of things, my overall sympathies lie more with the neoplatonic “strictly more than fact” perspective, and even more so with Aristotle’s more modest view that the “First” cause is strictly a final cause.

Hume on Causes

The great British empiricist philosopher David Hume (1711-1776) wrote that “There are no ideas which occur in metaphysics more obscure and uncertain than those of ‘power’, ‘force’, ‘energy’, or ‘necessary connection'” (An Inquiry Concerning Human Understanding, Library of Liberal Arts ed., p. 73).

Hume is most famous for his critique of naive or dogmatic assumptions about causality. “[T]here is not, in any particular instance of cause and effect, anything which can suggest the idea of power and necessary connection” (p. 75). To me, it seems to be the idea of an underlying power or force responsible for causality that he is mainly questioning. He has no doubt that we continually experience instances of cause and effect.

“[There is] no such thing as chance in the world” (p. 69). “[T]he conjunction between motives and voluntary actions is as regular and uniform as that between the cause and effect in any part of nature” (p. 98). Clearly, then, he did believe in the reality of cause and effect, but only wanted to reject naive claims about our knowledge of the world that purport to link experienced instances of cause and effect to explanations in terms of the operation of underlying powers. What we actually experience in these cases is just lawful regularity.

“It is certain that the easy and obvious philosophy will always, with the generality of mankind, have the preference above the accurate and abstruse…. The fame of Cicero flourishes at present, but that of Aristotle is utterly decayed” (pp. 16-17). But on the other hand, “All polite letters are nothing but pictures of human life in various attitudes and situations…. An artist must be better qualified to succeed in this undertaking who, besides a delicate taste and a quick apprehension, possesses an accurate knowledge of the internal fabric, the operations of the understanding, the workings of the passions…. However painful soever this inward search or inquiry may appear, it becomes in some manner requisite to those who would describe with success the obvious and outward appearances of life and manners…. Accuracy is, in every case, advantageous to beauty, and just reasoning to delicate sentiment. In vain would we exalt the one by depreciating the other” (p. 19). Clearly, then, Hume’s polemic against scholastic modes of reasoning does not at all mean he simply rejects the values of “accurate and abstruse” philosophy.

Neoplatonizing tendencies in the Aristotelian commentary tradition led to the common Latin scholastic view of causes as metaphysical powers operating behind the scenes that Hume is mainly concerned to criticize. Aristotle himself identified causes more broadly and much less speculatively with every kind of “reasons why” things are as they are and behave as they do. He did so without making extravagant claims to certain knowledge. Whereas scholastic philosophers characteristically debated the pros and cons of accepting various abstract propositions, Aristotle himself was fundamentally concerned with the use of reason to help interpret concrete human experience (see Aristotelian Causes).

Hume is a great philosopher, and so far I have focused on a positive appropriation of his work, having some points in common with themes I have been pursuing about causality and the notion of power. Robert Brandom’s innovative reading of Kant’s response to Hume points out that there are distinct limits to Hume’s approach.

“Kant read Hume’s practical and theoretical philosophies as raising variants of a single question. On the side of practical reasoning, Hume asks what our warrant is for moving from descriptions of how things are to prescriptions of how they ought to be. How can we rationally justify the move from ‘is’ to ‘ought’? On the side of theoretical reasoning, Hume asks what our warrant is for moving from descriptions of what in fact happens to characterizations of what must happen and what could not happen…. Hume’s predicament is that he finds that even his best understanding of facts doesn’t yield an understanding of either of the two sorts of rules governing and relating those facts, underwriting assessments of which of the things that actually happen (all he thought we can directly experience) ought to happen (are normatively necessary) or must happen (are naturally necessary).”

“Kant’s response to the proposed predicament is that we cannot be in the situation that Hume envisages: understanding matter-of-factual empirical claims perfectly well, but having no idea what is meant by modal or normative ones” (Brandom, Reason in Philosophy, p. 54).

Brandom continues, “To judge, claim, or believe that the cat is on the mat, one must have at least a minimal practical ability to sort material inferences in which that content is involved (as premise or conclusion) into good ones and bad ones, and to discriminate what is from what is not materially incompatible with it. Part of doing that is associating with those inferences ranges of counterfactual robustness…. So, for example, one must have such dispositions as to treat the cat’s being on the mat as compatible with a nearby tree being somewhat nearer, or the temperature a few degrees higher, but not with the sun being as close as the tree or the temperature being thousands of degrees higher. One must know such things as that the cat might chase a mouse or flee from a dog, but that the mat can do neither, and that the mat would remain essentially as it is if one jumped up and down on it or beat it with a stick, while the cat would not. It is not that there is any one of the counterfactual inferences I have mentioned that is necessary for understanding what it is for the cat to be on the mat. But if one makes no distinctions of this sort — treats the possibility of the cat’s jumping off the mat or yawning as on a par with is sprouting wings and starting to fly, or suddenly becoming microscopically small; does not at all distinguish between what can and cannot happen to the cat and what can and cannot happen to the mat — then one does not count as understanding the claim well enough to endorse it” (pp. 54-55).

Brandom concludes, “If that is right, then in being able to employ concepts such as cat and mat in ordinary empirical descriptive claims one already knows how to do everything one needs to know how to do in order to deploy concepts such as possible and necessary — albeit fallibly and imperfectly” (p. 55).

I am actually a little more sympathetic to Hume, in that I don’t read him as categorically rejecting the validity of concepts of necessity, only any possibility of certain knowledge of how they apply to the real world. I personally like the position of Leibnitz that necessity is real but always hypothetical, never categorical. But Brandom is right that Hume does not go on to emphasize how essential our fallible understanding of necessity is to our understanding of ordinary experience.