Hume's Two Definitions of Subjective Causation

I. Introduction

One of the oddities of Hume’s texts is his presentation of two separate definitions of causation, taking each the form of a natural and philosophical relation. He is also considered to rank highly among the tradition of skeptics, wherein all categories of things including causes are often read as fictions. We might wonder if we are to be skeptical enough regarding the limits of what we can say about causation in the first place, how are we to motivate a substantial difference separating the two definitions? Both forms of causation would hinge on the capability of demonstrating any kind of causation at all which Hume regards as, “One of the most sublime questions in philosophy” (Hume 112). This paper aims to explore the relation between the two definitions through a number of relevant examples and the contributions of the contemporary literature.

II. The Billiard Game

Hume presents a thought experiment wherein we are supposed to predict what will take place when we hit a billiard ball. Somebody who is an expert at the game has many memories available to themselves of past experiences playing. For Hume, these memories are impressions of sense equally so as any immediate perception that may take place of balls actually colliding. Sense perceptions can be differentiated by the force of their impression as compared with memories, or the imagination. We can iterate over these past experiences and understand many similarities occurring between these cases of balls colliding in what may be seen as ‘families’ of types of events taking place. Likewise, when actual balls are right now seen colliding, they may conform to these types as well. In this way, we develop an intuition of causation in the form of the natural relation and as Hume notes, “…all this is independent of and antecedent to the operations of the understanding” (Hume 121). This works to group up objects we have experienced which share the feature of being conjoined in such a way, and also at once the way we take part in this grouping up of them. It is only our own insight that they have this feature that allows an understanding of the relation of natural causation.

However in this way, causation cannot be seen as playing itself an active force in the world. We have no impression of power at work definitely causing something to happen, there is just the apparent sequence of moments. If it is necessarily only the way we happen to view it, we may be wrong in each one of our instances of attributing causation to an event, or that the world is even characterized by causation in any real sense whatsoever. We should see then that, “An object may be contiguous and prior to another, without being considered as its cause. There is a necessary connection to be taken into consideration; and that relation is of much greater importance” (Hume 56). There would be no use of this sort of cataloguing of past events or even presently occurring ones if we could not go beyond that definite knowledge and produce some kind of a relation with the future. We often hope for this relation to be predictive, in a probable sense. So, the more times events are seen as occurring in a conjoined way, the more likely they are to happen again - and specially when the preceding type of object is what happens to be currently perceived.

The billiard game only has meaning if we assume that the next case of hitting a ball will be a representative case of this population¹. The natural relation of causation exactly has meaning in this sense, but it necessarily cannot have meaning without making this kind of a contingent assumption about the possible nature of the next case. Hume refers to this as a ,"…secret tie or union among particular ideas, which causes the mind to conjoin them more frequently, and makes the one, upon its appearance, introduce the other" (Hume 10). Many empiricists, who see themselves as following in a certain tradition of Hume deliberately assume this subjective mode of probabilistic thinking as one of the shining virtues of their practice. The definition of causation we have been working with up until now then reads in Hume exactly as follows: “A cause is an object precedent and contiguous to another, and so united with it, that the idea of the one determines the mind to form the idea of the other; and the impression of the one to form a more lively idea of the other” (Hume 122). What the meaning of these frequencies is besides that they are seen to exist cannot be said. To put this another way, no number of past experiences of playing billiards should lead to definite knowledge about such causation. We may only arrive at different extents of an understanding regarding these statistics, namely by accumulating more and more cases, and not go beyond them. Players will continue to boast of certainty, and some will often appear to be correct.

Viewed as a radical empiricist in one light, we may wonder what reason Hume could have had for ever believing that there should be any regularity in life. From a perspective of total skepticism, there should be neither regularity nor experience in the first place. Each one of our of experiences is suspect in itself, and none of them should result in the emergence of an idea of ’experience’. What could really be said inside an experience is that some impression of sensation was had, and there would not be the time to say it.

For such an empiricist, everything can and even must be retraced back to its origin in experience, including most importantly all ideas. No idea can be said to have its right of origin from anything other than an initial impression, or another idea which should still itself end in impression. Hume says of all our simple ideas in particular that they, “…in their first appearance are derived from simple impressions, which are correspondent to them, and which they exactly represent” (Hume 3). What is it about this conjunction which happens with some frequency and seems to even occur involuntarily? Such a connection should follow were one term to have brought the other about by directly causing it to happen. However, if causation does is not seen in this way to have taken substantial form, why should we say that the impression necessarily leads to the idea and not the other way around? We have only assumed that the constant conjunction is actually such as it appears, and so we could not provide a necessary demonstration no matter how obvious it would appear to be sufficiently so.

There is a dichotomy which is particularly acute in much of Hume’s work that holds between the scientific principles of naturalism and empiricism. Hume’s naturalism occurs in his language regarding the collection of evidence resulting in probable knowledge. It unifies all causation in a pre-discursive sense, as everything exists in one shared, unified world, and nothing is transcendent of that. Though we distinguish between natural and philosophical forms of causation, it is only a formal distinction being made about an ostensibly monistic world. We can make all kinds of caveats about the matter, and that in itself is why we are so careful to make the distinction that naturalism is an important assumption that grounds our capacity for sharing complex claims with each other. However, each one of our experiential claims is also suspect in itself, “It may, therefore, be a subject worthy of curiosity, to enquire what is the nature of that evidence which assures us of any real existence and matter of fact, beyond the present testimony of our senses, or the records of our memory” (Hume 19). An aggravating skepticism might undermine each and every one of these claims.

We must distinguish in Hume between relations of ideas and matters of fact. Some of the relations of ideas are causal relations. Certain connecting principles can be seen to motivate the formation of natural relations of causation, having to do with historical frequencies of causes being conjoined with events. But the matters of fact themselves and the way philosophical relations become freely conjoined each cannot be explained in this way. The formations of associative relations which hold between these make up different kinds of rules and regularities, but they are also subject to negation through contrariety, since, “The contrary of every matter of fact is still possible, because it can never imply a contradiction…” (Hume 18). Hume proposes that we begin with how we actively associate certain ideas together in the mind as possessing an apparently necessary connection, and explore this form of belief - even if it is taken purely in the restrictive hypothetical sense as being a fiction. A primary association we tend to find is the causal one, and we see that it can be either of a philosophical or natural type. It would seem to follow that the natural causes are the obvious foundation for our theory, preceding the philosophical ones. Contrariety is only found in the philosophical relation however, and so given only the natural relation a difference could not be supposed.

It may be a total absurdity to abuse the notion of ‘fiction’ as the foundation of an entire philosophy. Even other radical empiricisms attempt not to undermine themselves right off the bat in such a way, being rather more ambiguous in their first formal principles. Where could Hume have been going with this idea? In an initial attack on innate rationalism, he declares that, “All ideas, especially abstract ones, are naturally faint and obscure: the mind has but a slender hold of them: they are apt to be confounded with other resembling ideas” (Hume 15). This should be read as a naturalist’s demand to subject each and every one of our beliefs to the potential force of critical measure. Viewing ideas as fictions allows us to treat them in a manner which could lead to our personally-invested belief in each one of them being found to have been arrived at in error. Whereas distinctively for the player, there comes a time in the game when the decision to hit the ball must be made based on whichever set of factors and outweighs the desire for any further speculative quandary as to what might be the case. Here we find the necessity of the natural relation of causation definitely at work².

III. Fictions and Probabilities

Let us turn to a speculative point. Hume tells us that, “Our conclusions from experience carry us beyond our memory and senses, and assure us of matters of fact which happened in the most distant places and most remote ages” (Hume 34). Are all parts of these most distant places and most remote ages merely subjective fictions? The entirety of our access to them may conceivably at least be characterized as such. It would probably be implausible to be skeptical about all parts of the world and its apparent history all at once, but Hume does want us to have the liberty to call into question any given part of it - particularly in the case that it were to have some reason for raising our suspicions. In order to get the grounds for discourse under way, Hume would probably have to admit to being a realist about many things and much of history that is not presently the object of scrutiny. The difference between philosophical causation and natural causation is mostly that when custom takes the strongest form, “…it not only covers our natural ignorance, but even conceals itself, and seems not to take place, merely because it is found in the highest degree” (Hume 21). This poses an obligation for philosophy to root out these customs of ‘highest degree’, so that they too may be subjected to critique.

A problem arises though that we cannot reason about what beliefs are. Any belief may have originated from fiction and so this leaves Hume with a difficulty in distinguishing them. He could swindle us with one story or another about something only illusory underlying their difference, but we would want to know further whether the difference is real from a perspective outside of our own partial judgment. What makes a belief different from a fiction cannot be some substantial qualitative property inherent in the belief which decisively establishes it as ’non-fictional’, there is certainly no such demonstrative proof available. They are very closely related then, but a certain sentiment accompanying the senses and memories inflects our beliefs and separates them from fictions, which, ‘…renders realities … more present to us than fictions, causes them to weigh more in the thought, and gives them a superior influence on the passions and imagination" (Hume 37).

So much of the memory of events are qualitatively different from the pure machinations of our imagination, we may initially observe that belief is susceptible to the force of habit. Beliefs can be taken for particular kinds of relations in themselves on the basis of the accompanying sentiment and not always purely in unison as fictions. In fact we could feverishly end up being wrong in each one of our accounts concerning beliefs and fictions but this would change nothing about the difference in accompanying sentiment or lack thereof. It might be a bodily pain which is presently severe in the immediate moment, or a conjunction of events which is seen so frequently as to be nearly assured. As Hume argues, “…there is a considerable difference between the perceptions of the mind, when a man feels the pain of excessive heat, or the pleasure of moderate warmth, and when he afterwards recalls to his memory this sensation, or anticipates it by his imagination” (Hume 11). In the force of belief, we find a kind of causation already at work which allows for further inquiry to be made.

Even when we have managed to trace an idea all the way back to its impression as sense perception, Hume argues, “…their ultimate cause is, in my opinion, perfectly inexplicable by human reason, and ’twill always be impossible to decide with certainty, whether they arise immediately from the object, or are produced by the creative power of the mind, or are derived from the author of our being” (Hume 61). The transition for the most part is seen to take place spontaneously, but if it can be said to be of anything further it is of causation first and not only of reason. Impressions absolutely cannot arise purely out of reason, “The inference or transition from impression to idea does not arise from experience through reason, for that would require the principle of the uniformity of nature … which is provable neither demonstratively, nor probably…" (Selby-Bigge 648). Are the principles of association which produce the sense of causation completely random then? Hume says they cannot be, “…were ideas entirely loose and unconnected, chance alone would join them” (Hume 10).

What he tells us of chance is that, “There be no such thing as chance in the world” (Hume 43). But more interestingly, what he says of probabilities is, “The case is the same with the probability of causes as with that of chance” (43). Chance is not of nature because the existence of things is not doubtable necessarily in the way that our ideas about them are. In fact, each thing either definitely exists or does not exist and there can be no in between. Hume says of chance that it is, “…nothing real in itself, and, properly speaking, is merely the negation of a cause, its influence on the mind is contrary to that of causation” (Hume 90). The regularities in the mind are generally more representative (and this is both of sensations and of reflective impressions and ideas), than they are purely imaginative. All of our imaginations would seem to amount to a complex bundle which begins with the simple impressions we first encounter. We should see then that chance must, “…leave the imagination perfectly indifferent, either to consider the existence or non-existence of that object, which is regarded as contingent” (90).

Change is expected to happen because of a latent principle in the world which will bring about said change. But other philosophical relations which evince a self-certitude must have to do with a future actually becoming brought about freely. Now we must proceed to a different but related notion of causation which states in a perhaps confusingly similar fashion, “An object precedent and contiguous to another, and where all the objects resembling the former are placed in like relations of precedency and contiguity to those objects, that resemble the latter” (Hume 122). This notion is most importantly explicitly subjective, as it entails an implicit expression of causation which happens in an unspecified material, which is presumably still in any case going to have to be inside the mind.

This is to say that being a cause is not a quality inherent in things but a relation that is seen to hold between them, i.e. the experience of one event following another. It is not then the case that each thing must have an individual cause, we may experience a thing immediately as without having had a prior noticeable cause. Or we may also experience some thing as bringing about multiple effects, and so particular numbers of perceived causally-related things do not map well onto probabilities. Hume takes his subjective argument so far as to suggest that, “Any thing may produce anything” (Hume 124). These points would seem to entail a skeptical refutation of induction. He concedes at least that, “…one proposition may justly be inferred from the other; I know, in fact, that it always is inferred … The connection between these propositions is not intuitive” (Hume 25).

Subjective skepticism potentially erases all distinctions that were made about what we thought was the physical world. This is a problem which Hume wrestled over incessantly and mostly did not overcome in his own mind. What needs to be defended is only a certain kind of speculative belief, “…unless we should entertain such a scepticism as is entirely subversive of all speculation” (Hume 8). It is one of the difficulties certainly in mining his work for positive theoretical content. It has often been read as exhibiting very destructive tendencies. Selby-Bigge calls the natural-philosophical causal relation distinction itself an “invidious contrast,” having to do with the one’s skepticism-inducing subjectivity over the other (Hume xvii). Hume says this skepticism is mistaken where it leads, “…even to the absolute rejecting of all profound reasonings, or what is commonly called metaphysics” (Hume 4).

What are we to make of this conclusion in terms of probabilities? No frequency of representation in experience should amount to more formidable certainty that a past event has necessarily taken place as imagined. Hume has a similar argument, “From the mere repetition of any past impression, even to infinity, there never will arise any new original idea, such as that of a necessary causation, and the number of impressions has in this case no more effect than if we confined ourselves to one only” (Hume 63). But this is exactly contrary to what is expected in a frequentialist understanding of the future arising directly out of some field of possibilities, even provided the understanding that real underlying frequencies of possibilities are necessarily inaccessible to us. There is a question here of how to surmount the problem of frequentialism, or as Johnston asks (2011),

If material being an sich is contingent qua containing within itself no law-like necessary connections, then why isn’t reality and the experience of it a violently anarchic and frenetic flux? Asked differently, how come there are apparently stable causal orders and structures if absolute being actually is hyper-chaotic?

The answer is nothing, there is no reason why the world is the way that it is. But we do not expect them to change, what they are very precisely is the sense that in certain ways the world has been characterized by these very structures remaining the same at least until for right up until now. Meillassoux argues, “Yes, we do hold that the laws of nature can effectively change without reason; yet, we do not expect them to change any time soon, anymore than anybody” (Meillassoux 114).

IV. Contingency

We have found that Hume is an early subjectivist about causation and in his general philosophical tendency. In this sense, he follows in a tradition with Descartes and Berkeley. Glassford argues though, he is also a skeptic (or even a subversive) about innate rationality and its obvious connections with authoritarianism, “Hume reported that while writing the Treatise he was seized by a desire not to submit to any authority, a youthful rebellion which certainly captured something of what postmodern philosophers have suggested they are about today” (Glassford 2). Going further, “…for Hume the ultimate nature of reality was always a fiction of the imagination, never an object of the understanding, as it was for Kant” (2). Hume is defending a philosophy in which speculative belief has any effect on action whatsoever, weaving a careful line between what often results for other philosophies in either a dehumanizing determinism or an ineffectual relativism. importantly therefore, he levels the following argument against a principle of sufficient reason:

We have already taken notice of certain relations, which make us pass from one object to another, even though there be no reason to determine us to that transition; and this we may establish for a general rule, that wherever the mind constantly and uniformly makes a transition without any reason, it is influenced by these relations. … which associate together the ideas of these objects, and unite them in the imagination (Hume 66).

It is better to say that every event has a cause than that every event has a reason. Of course we could say that we could only know of an event’s cause by way of reason, and so the only events we would ever know of are event’s that have reasons we could give for them, but Hume says that we first have acquaintance with conjoined events by way of simple impression, in the form of immediate perceptions, and only after this are we able to form any fully fledged ideas about them. Hume complains about abstract ideas that, “…the chief obstacle … to our improvement in the moral or metaphysical sciences is the obscurity of the ideas, and ambiguity of the terms” (Hume 47). We must be careful to, “…pass from words to the true and real subject of the controversy,” as ideas often get confused for a substantial determination merely directly referred to by some word (Hume 284).

As soon as one writes a contingent claim the possibilities of its erasure by future negation are already written on to it within the span of the numerical, qualitative differences which hold for the relations constructed on its group of terms. Because, “…nature, by an absolute and uncontrollable necessity has determined us to judge,” judgment does not conform to the truthhood of the proposition it governs or impossibly a significant lack thereof, but to the merely empirical matter of fact of its own existence (Hume 131). Any discussion involving ontology is not essentially too abstract or we are espousing a quietism and leaving a degenerate mode of metaphysics in place, people are allowed to say what they think are metaphysical ideas, but the philosopher will always be there to step in and suggest that only what we are talking about is not actually metaphysics like we suppose that it is. Let’s move on now to what we should speak about rather than simply what we may or may not be able to.

This argument eliminates the necessity of any and every particular reason, “…it consists in the thought of thought‘s contingency for the world, and the recognition that thought has become able to think a world that can dispense with thought” (Meillassoux 116). Hume himself is entirely opposed to dogmatic metaphysics, decrying, “In vain do we hope, that men, from frequent disappointment, will at last abandon such airy sciences…” (Hume 7). That time and time again we fail to uncover the necessary reason for why the world is as such has rarely led to the conclusion that there is not one at all. There is no use for Hume as a philosopher in discovering a natural causation and discovering a philosophical causation as an absolute, substantial difference. Hume found the advances being made in natural science of his day very exciting, so much so that they were deserving of a demonstration for their justification. Deleuze explains of this method, “Human nature means that what is universal or constant in the human mind is never one idea or another as a term but only the ways of passing from one particular idea to another … And yet at first Hume’s thesis seems disappointing: what is the advantage of explaining relations by principles of human nature, which are principles of association that seem just another way of designating relations?” (Deleuze 39). In this way though, we make work with even purely subjective experience as the material in an investigation of causes, and Hume shows how this is at all possible.

Hume must be seen as a skeptic concerning the definitional meanings of individual words, as he is interested in how they can be employed in and alongside a constellation of other already existing meanings. It would be very difficult for him to define the difference between two words without quickly resulting in circularities and remapping of definitions based on prior partial formulations. Where the difference between natural and philosophical cause has any import, it is only through a dialectical analysis of their placement within a particular context. We can surmise that is usually going to entail a concern with shoring up some form of absolute certainty - and Hume’s whole project is a rejection of this form of philosophy. Natural and philosophical causation are not each in themselves, but are their mutually interdependent relation to one another and also to all of the other terms at work at once in Hume’s theory.

We have committed all speculative reasoning (a priori, metaphysical) to flames, but what of speculation in itself, for example that which always tends to be involved in a preliminary stage of the acting out of the passions? Meillassoux says even further and much more strikingly of mathematics that, “…it is possible to envisage an absolutizing thought that would not be absolutist” (Meillassoux 34). This is not in the least a skeptical position, as Gironi adds, “This speculation proceeds via rational demonstration, since the absence of reason does not entail the end of rationality; on the contrary, it is rational thought that leads reason to the liquidation of necessity” (Gironi 36).

Hume expresses the relevant problem such that, “There is required a medium, which may enable the mind to draw such an inference, if indeed it be drawn by reasoning and argument. What that medium is, I must confess, passes my comprehension” (Hume 47). The medium Hume is searching for is not any infinite excavation into the depths of the field of possibility, which admit of no end, but rather a matter of contingency. The natural relation of causation may be said to be governed by various probabilities, and whether or not those map on to some sense of what happen to be the correct ‘objective’ probabilities. We will examine the distinction not as an unbridgeable gap holding between a now defunct subject and object, but rather as illuminating that of the empirical and the absolute in contingency. The philosophical relation exposes another domain of contingency, through which the actual freedom of the causal role played by the imagination becomes realized. The medium of contingency could speculatively be said to be material, transactional, quantitatively differential, without merely consisting in sets of possible worlds.

Ayache differentiates them so that, “In the real world, contingency is not reducible to underlying states because truly unpredictable events are typically those that escape the previously known range of possibilities or possible states” (Ayache 40). This is not even to explicitly falsify any given range of possibilities but only to say that they are each relevant to their own context and not outside of it. The extent of the distancing is such that, “The current saying is that true events create the possibilities that will have led to them. It is out of the question to apply probability to those events, or even to say that their probability was zero” (40).

Empirical contingency characterizes both natural relations and philosophical relations of these past sequences of causation. It concerns specifically actual matters of fact about the past, and the counting of certain kinds of related causes and events. But other causal relations which evince a self-certitude can only be rather of an absolute contingency, which must have to do with a future actually becoming brought about. These reveal that freely, anything may change and always at once that nothing may happen. It is only the actual becoming into reality as a present moment of some virtual future that will have brought about the motion of a real event. Both natural and philosophical relations of causation can be reasonably governed by the form of empirical contingency. But absolute contingency can only be arrived at from philosophical relations of contingency, because it cannot arise from any laws characterizing a set of already existing possibilities that constitutes the law of association at work in natural relations.

Can we think about whether the movement of the billiard ball could have happened otherwise, such that it veered only slightly to the side of its actual direction? What relation does this form of contrariety in the imagination share with the event? Presumably, somebody who possesses a very good intuition about a large number of alternatives will be able to readily select the best from among them, and in this way expert players usually are able to hit their shot. Real odds are nothing other than imagining it actually having happened otherwise, they are not out there in the world waiting for us to arrive at them, rather they arise directly out of the labor of contrariety in imagining the possible negation of causal histories. While all causal histories are potentially subject to negation³, you do not get this for free. Some histories or parts of histories may seem unassailable, such as the present moment or the distant past. But they too are open to at least possible negation.

Absolute contingency is opposed to both possibility and necessity; we are not dealing merely with expectation of a particular future but moreover its being brought about in an action through an associated belief. Ayache says of the absolute, “The outside it is pointing to is not the outside of the other possible worlds, but the outside of thought … it subtracts the condition of being thought … The absolute is therefore even simpler than the empirical, as it is not even opposed to metaphysics; it ignores metaphysics” (Ayache 202). In natural relations of causation, we find a connecting principle which eventually must involve the specification of ranges of possibilities. It could not in itself express the actual force of the real becoming of a future, which does not merely consist in the labor of listing however many different possibilities⁴. In philosophical relations of causation, the cause and the event are seen to be freely conjoined. This can only be a matter of a future becoming into being in which the connecting principle that the ideas conform to could not have determined them to be seen to do so necessarily, otherwise they would simply always have been so already - everything would immediately be found through tautology. All of the philosophical relations of causation that are seen to have been freely conjoined in the imagination may indeed have actually been produced unconsciously by another connecting principle. However, a real cause bringing about of an event of the future is not simply contained in the list of possibilities already existing from the present moment. It is also not brought about merely by not being on this list. Contrariety does not result in a chaotic infinite of events otherwise than what exists becoming at all times. It could have not been brought about at all, and most of the time when philosophical relations are not deliberately employed this is exactly what we would expect to happen.

How may we continue to do philosophy without there being necessity to any of the laws of nature? Many before Hume had skeptically denied such necessity, but in the same rhetorical fell swoop went any hope for a coherent philosophy. Instead, what we find is that necessity arises out of expectation, it is, “…only a strong and steady conception of any idea, and as such approaches in some measure to an immediate impression,” but it does so contingently (Hume 70). Because of this, all causation must be seen as subjective.

References

Ayache, Elie. The End of Probability. Wilmott.

Ayache, Elie. (2010). The Blank Swan: The End of Probability. Chichester, West Sussex, U.K.: John Wiley & Sons.

Deleuze, G., & Boyman, A. (2001). Pure immanence: Essays on a life. New York: Zone Books.

Gironi, Fabio. (2011). Meillassoux’s Speculative Philosophy of Science: Contingency and Mathematics. Pli.

Glassford, J. (2008). Disrupting Philosophical Boundaries: Gilles Deleuze on David Hume.

Hume, D. (2004). An enquiry concerning human understanding: And selections from A treatise of human nature ; with Hume’s autobiography and a letter from Adam Smith. New York, N.Y.: Barnes & Noble.

Hume, David, L. A. Selby-Bigge, and P. H. Nidditch. (1978). A Treatise of Human Nature. Oxford: Clarendon.

Hume, D. (2003). A treatise of human nature (Dover ed.). Mineola, N.Y.: Dover Publications.

Johnston, A. (2011). Hume’s Revenge, À Dieu, Meillassoux?. re.press.

Laing, B. M. “An Abstract of A Treatise of Human Nature, 1740. By David Hume. Cambridge at the University Press. 1938. Pp. Xxxii 32. Price 3s. 6d. Net.).” Philosophy 14.53 (1939): 116.

Meillassoux, Q. (2009*). After finitude an essay on the necessity of contingency*. London: Continuum.

Notes