Logical and Spiritual REFLECTIONS
Book 1. Hume’s Problems with Induction
Chapter 5. The self or soul
As we saw in the examples of Hume’s psychological theories of generalization as habit and of causation as association of ideas, he tended in practice to engage in faulty induction (and of course, faulty deduction).
He synthesized from a little data or a superficial analysis, without paying heed to information or arguments that would have delimited or belied his foregone conclusions. He would focus on or select positive aspects of an issue, those that confirmed his theses, and blithely ignore or discard negative aspects, those that weakened his positions.
Such faulty practices on his part are not surprising, in view of his theoretical opposition to induction, i.e. his belief that induction has an intrinsic problem. If one has a general failure of logical understanding, this will inevitably eventually translate into errors of practice. Conversely, the theoretical error is itself due a practical failure. Of course, such error is never ubiquitous; else the person committing it could not at all engage in discourse.
The same tendency of faulty induction is to be found in Hume’s treatment of the human soul and of freedom of the will. Rejecting offhand the Cartesian inference “cogito, ergo sum”, Hume denied the existence or knowability of a human self or soul, conceiving our common belief in such a thing as due to nothing but the “bundling or collection” of our various perceptions:
“It must be some one impression, that gives rise to every real idea. But self or person is not any one impression, but that to which our several impressions and ideas are supposed to have a reference. If any impression gives rise to the idea of self, that impression must continue invariably the same, through the whole course of our lives; since self is supposed to exist after that manner. But there is no impression constant and invariable. Pain and pleasure, grief and joy, passions and sensations succeed each other, and never all exist at the same time. It cannot, therefore, be from any of these impressions, or from any other, that the idea of self is derived; and consequently there is no such idea.”
Though his thinking on this important issue, as on many others, is clearly based on personal observation and insight, showing Hume to be a real philosopher, worthy of considerable respect, his reasoning is here again faulty. He argues that we would need to experience a single “impression”, one permeating our whole experience, to justify the idea of a self. By this, he seems to mean a concrete mental phenomenon of some distinct sort. Not finding such a core experience, he reduces our personal identity to at best the sum total of the mass of fleeting impressions of all sorts that we obviously have. But we may disagree with this viewpoint on several counts.
First, on what ground does Hume demand at the outset that the self be configured in the way of a single permanent “impression” underlying all inner experience? That must be seen to be a hypothesis of his, one that needs to be inductively proven, and not necessarily as he assumes the only possible way of conceiving the issue. The self might not be as phenomenal an entity as he projects (i.e. an impression), and it may be wiser to define it by referring to its functions (cognition, volition and valuation) rather than to its substance.
With regard to Hume’s condition of singularity of impression: it would not be inductively erroneous to claim that the self is the sum total of all impressions. This might be taken to mean that all our impressions are indicative of or even actually cause an underlying entity, which though never perceptible is assumed to endure through time. In other words, the whole is more than the parts. Such assumption would simply constitute a conceptual hypothesis, like for example the hypothesis of electrons in physics as entities underlying electrical phenomena. An abstraction does not have to be identical with the experiential data that supports it.
With regard to Hume’s condition of permanence of impression: to demand as he does that we be conscious of the self full time, or even part time, before we believe in it, is not in accord with inductive logic. The latter allows us to extrapolate from occasional apparent self-awareness to an assumption of permanent presence of a real self – this would just be generalization. We might even postulate a self without any direct impression of it, in the way of an adductive hypothesis to be supported by various other experiences and considerations. Either approach would be in accord with inductive logic, provided we obeyed the usual rules of induction (especially, that no contrary evidence or inconsistency be found).
Secondly, Hume is arguing in a circular manner when he says (in the above quotation): “It cannot, therefore, be from any of these impressions, or from any other, that the idea of self is derived”. Even if we accepted (which I do not, as just explained) his contention that the self cannot be inferred from impressions other than that of the self, it does not follow that we do not in fact have impressions of the self. When he says “or from any other”, he means to categorically exclude this special experience, which he claims never to have.
We need to seriously consider the empirical and inductive status of Hume’s claim to have no self-awareness. It is important to note that this claim is negative, which means that it reports an unsuccessful search for something (an impression he can identify with the self). How much introspective observation is this claim actually based on? Did he meditate with great effort an hour a day for five years, say, in search of his self? Or did he, as I suspect, casually look into his mind for five seconds of so, a couple of times, and conclude what he had already decided to assert as true, viz. that he had no self?
Moreover, whether proposed prejudicially or casually, or after very conscientious investigation, a negative statement like that always and necessarily involves a generalization. We generalize from “I looked everywhere in me for a long time, and did not find what I sought” to “there’s no such thing as the thing sought, in me or anyone else”. This to repeat is a generalization, and there is no way for us to arrive at an empirical negative statement in any other way.
Hume generalizes: from the few moments when he perceived no self, to all his temporal existence; and from his own inner life to the same condition in all other persons. Yet Hume does not officially believe in generalization! Is he exempt? Are we to suppose that he is allowed to generalize (and indeed to do so from very tenuous data, his doubtful introspection), but no one else is? This is clearly either a double standard or a self-contradiction on Hume’s part. He postures as an empiricist, and is widely so regarded, but his empiricism is clearly very superficial and make-believe.
Thirdly, there is an alternative position (which I adhere to), which is fully in accord with the principles of inductive logic. It is that we all do experience our own self quite often, though such experience may vary in degree and depend on circumstances. The self is always implied and present, in every moment of cognition, volition or valuation. But to be aware of it, or sufficiently aware of it to declare it present with surety, an effort of ‘self-consciousness’ is needed.
Moreover, such self-consciousness is not a perception, but an intuition, because the self is not a phenomenal entity (i.e. one with visible, audible, or other sensible qualities), but a non-phenomenal one. To experience it, one must aim one’s awareness ‘inward’, i.e. towards the sought-for subject, and not outward in the direction of mental or physical objects.
A lot of meditation practice is needed to pacify, silence and still the mind sufficiently to contemplate the self with some clarity and confidence. If there is a stage at which the self effectively disappears, or is seen to be ‘empty’, as some advanced meditators claim, that stage is much deeper than Hume ever evidently went. So Eastern philosophy cannot be appealed to in support of Hume.
If one expects to find the self in gross sensory or mental “impressions” of the sort Hume had in mind, one will of course be disappointed. But if one realizes that the self is a much more subtle appearance than those, to be apperceived rather than perceived, one can well claim to experience the soul directly.
It appears more readily in the way of a ‘presence’ inherent in all intentions and acts of consciousness, will and valuing, than as an isolated object. But there are suggestions that, at a deeper level, the self can be contemplated ‘in itself’, and further on (more mystically) as a part or aspect of a universal Self.
Additionally, we have a justifiable concept of self. We could accept the self as no more than a conceptual construct – this would logically be an acceptable position. We are logically allowed and even recommended to propose hypotheses that unify and explain empirical data.
We could well argue that events like consciousness, volition and valuation imply a self. They are incomprehensible without the assumption of a self. To be conscious is to have a self; to will is to have a self; to desire or dislike is to have a self. The brain and other sense and motor organs are not themselves conscious or in possession of the power of will; these are not a subject or agent, but mere channels or instruments.
But in my view, this narrow, constructivist position would not explain all the facts of experience. For how would we then claim to know specific particulars about our own individual mental workings from such a general abstraction? To overcome this difficulty, we have to adhere to an intuitionist postulate.
For instance, if I have a thought right now, I can intimately tell whether that thought is my own will, or occurring without or against my will. I am quite able to distinguish between my own beliefs, wills and values – and those imposed on me by my brain or external influences. If I had no direct intuition of myself, or at least of my own inner acts, no such distinction would be feasible.
No theoretical knowledge of the self can produce such intimate certainties. Therefore, we must admit we do experience the self itself – if only occasionally, e.g. when we specifically make the effort to do so. It is not merely a concept for us, but also a direct experience.
A difficulty in self-awareness is perhaps due to our inability, except possibly in deep meditation, to detect the self as such. Ordinarily, we experience our self through its actual functioning, i.e. when we are involved in particular acts of cognition, volition or valuation. When the self does not ‘express’ itself in any such acts, it is transparent like space is to our eyes, except perhaps (to repeat) in meditation. Although intuition of self is also an act of the self, there seems to be a requirement that the self first express itself otherwise than through intuition, before intuition can detect it!
Hume refused to acknowledge such appearances of self-consciousness as valid data. He engaged in introspection, but clearly not enough of it; perhaps he was too impatient, and drew a premature conclusion. He generalized – from his own non-experience of self at some time(s) to all persons forever. For these reasons, his negative conclusion cannot be considered an undeniable fact (as many take it to be). It is just a theory, one with very little and inconclusive evidence going for it.
For my part, I insist: there is non-phenomenal experiential data from which a concrete idea of self can legitimately be drawn. That momentary self can then be generalized and reasonably claimed more permanent, at least to the earthly lifetime of the individual. We can further speculate that the self exists before and after death; but that is another issue, much harder to establish inductively if at all.
We can furthermore, on the basis of the said subtle data as well as with reference to phenomenal impressions, adductively posit a concept of self, an abstract self. Such adduction is even possible without reference to the intuitive data, but merely on the basis of the grosser data that Hume acknowledges. The abstraction so begun then provides support for the intuitive data, and the intuitive data in turn serves to further confirm and enlarge the abstraction.
Thus, to conclude, Hume’s skeptical posture towards the self is mainly due to his personal difficulties with introspection and with inductive procedure. He sets wrong theoretical standards of observation and of judgment, and moreover fails in practice to adhere to his own rules and restrictions.
 Treatise, Book I, Part IV, Sect. VI.
 I do not mean to say that had Hume meditated sufficiently, he would necessarily have affirmed the self. Many presumably major meditators deny the self’s existence (e.g. the Buddhist anatman doctrine), or at least its knowability (e.g. in the Brihadaranyaka Upanishad: “Nobody can know the atman inasmuch as the atman is the knower of all things”) – not that I always agree with their logic. But the word of a casual observer like Hume is not comparable to that of such meditators. In any case, we are still faced with mere hearsay, which must be empirically and rationally weighed. The said meditators might well be right, but other people cannot take them on faith and abstain from meditation. To claim the knowledge for oneself, one must personally meditate like those meditators did. After that, one must also judge their theoretical claims, and not just assume they were infallible geniuses.
 Even as an extreme empiricist, in the sense of modern “logical positivism”.
 If we try to tell a blind man about color, he may ask us whether it is loud or smells nice or tastes good or feels rough. But we cannot answer his question with reference to such phenomenal qualities, because the answer is a completely other sort of experience. He may then say: there’s no such thing as color! But that is just because he cannot see it. Similarly, to experience the self, one needs to intuit it – one cannot perceive it, for it has no phenomenal characteristics.