Many readers may find it hard to determine if David Hume denied the existence of the self completely or if he simply had a radical view on it. After all, it would be almost impossible to deny the self. Yet this largely depends on definitions of the word “self”. So why can’t the self be constituted by the “several perceptions [which] successively make their appearance, pass, re-pass, glide away []” if they’re all a result of the same brain with its encoded memories? (Something that Hume, as an 18th century philosopher, would never have discussed.) Similarly, why should anyone believe that the self must be perceived as a single “impression” (as Hume argued against his opponents)? So it can be doubted that Hume believed in the no-self argument (associated with Buddhism as anattā or anātman). However, that would largely depend on definitions too.
Laypersons and Philosophers on the Self
On the one hand, there is the traditional philosophical view (which Hume argued against) that there’s a “fixed self” with a determinate identity. On the other hand, there are those philosophers who deny the self entirely. It can now be said that perhaps it’s possible that many laypersons have a position between these two extremes. Of course, this is hard to establish without empirical or scientific research.
First things first. When David Hume talked about what he believed philosophers and laypeople take to be the self, was he correct? Well, he might well have been correct when it came to various historical philosophers. What about laypersons? It can be argued that most laypersons don’t really have any explicit and/or detailed notion of the self. Alternatively, they don’t have the notion that Hume targeted.
Hume was actually talking about what he took philosophers to believe about the self. He made that clear when he told his readers that when it comes to “explain[ing] the nature of personal identity”, this subject had
“become so great a question in philosophy, especially of late years in England, where all the abstruser sciences are studied with a peculiar ardour and application”.
When I thought about the self (i.e., before I read any philosophy), it was all extremely vague and with next-to-no detail. All the self amounted to was… what? In retrospect, it’s really hard to say. Sure, it was something which was deemed to be continuous and, yes, somewhat fixed. Yet that couldn’t have been explained by my young… self. And neither can it be explained by many adults…
Perhaps that was partially Hume’s point.
However, the idea of a fixed substance that never changes is certainly not something I believed as a child, and I doubt that many adults believe that either. So perhaps this is the foisting of philosophical terms — and ways of thinking — on laypeople.
Why One Single Impression of the Self?
Take Hume’s idea that the self “must be some one impression that gives rise to every real idea”. Yet the “self or person is not any one impression”.
Who argues that laypeople rely on “one impression” to believe they are a self? Some philosophers might have done so, sure, but not most laypeople. So, again, perhaps Hume never even considered what the layperson believed on this matter.
What would an impression of a self look like? Did even Hume’s philosophical opponents believe that there is a single impression of the self which accounts for any belief in it?
Thus, rather than criticising philosophical accounts of the self, perhaps we should be criticising Hume’s empiricism. This means that Hume dismissed other accounts of the self because they didn’t abide by his own critical empiricism. He, like many philosophers, was hooked on terms and ideas from philosophy, and when anyone didn’t abide by them, then they were deemed to have flawed conceptions.
In order to buy into Hume’s (possible!) no-self argument, we’d need to believe that the existence of the self depends on a single impression. More fundamentally, what did Hume mean when he wrote that his opponents must “perceive something simple and continued, which he calls himself”? Here again, Hume was relying on a standard empiricist position. (That’s no surprise since he was an empiricist.) A layperson would only need to accept that the self is something simple, fixed and a continuant in order to fall foul of Hume’s arguments.
Again, a believer in the self certainly needn’t believe that the self needs to be perceived as being something that’s simple, fixed and a continuant. Indeed, it needn’t be perceived at all.
Of course, the obvious question here is: What is the (or an) alternative to Hume’s position?
As already stated, Hume was talking about philosophers. (He wrote about “setting aside some metaphysicians of this kind”.) So it may well be a waste of time discussing what the layperson may believe about the self. Indeed, it may even be a waste of time, in this context at least, discussing what scientists believe about this matter.
David Hume’s Alternative
Hume fixated on a single impression of any object when he wrote that “[i]t must be some one impression that gives rise to every real idea”. He, instead, argued that when it comes to what we take to be the self,
“several perceptions [which] successively make their appearance, pass, re-pass, glide away, and mingle in an infinite variety of postures and situations”.
Why can’t the self conform to Hume’s own description? After all, these several perceptions, etc. are occurring in the same mind or consciousness, some of them are recalled in memory, and they’re all a result of the same brain. (At least they have that in common.) Again, why can’t “successive perceptions” constitute the self? Of course, they can’t according to certain historical philosophical accounts. However, why do notions of the self need to abide by them?
Hume’s psychological empiricism is again stressed when he wrote that
“[w]e have a distinct idea of an object that remains invariable and uninterrupted through a supposed variation of time; and this idea we call that of *identity* or *sameness*”.
It can be doubted that laypersons really have such a view. This is Hume reacting against other philosophers. However, this means that it’s possible that some laypeople may have a more philosophically accurate view of the self than the philosophers Hume reacted against. That said, there’s no way I can know this other than via empirical research. And that situation was true of Hume too.
The Self and the Ship of Theseus
An entire section of Hume’s A Treatise of Human Nature states the Ship of Theseus paradox in various ways. Not only that: this idea was vital to Hume’s notion of the self and personal identity.
Hume himself explicitly mentioned a ship. However, he didn’t mention the ship of Theseus, nor Plutarch or Hobbes. (Hume would have known that they had already tackled this subject.) Hume wrote:
“A ship, of which a considerable part has been changed by frequent reparations, is still considered as the same; nor does the difference of the materials hinder us from ascribing an identity to it. The common end, in which the parts conspire, is the same under all their variations, and affords an easy transition of the imagination from one situation of the body to another.”
Hume had a problem with these accounts of the ship. Firstly, we can ask: The same… what? We can’t say “ship” because we’re supposed to be explaining why it is the same ship. (Stating that the ship is the same ship doesn’t give us much.)
The materials which make up Hume’s ship have changed many times. So its physical makeup can’t constitute its identity. So what does? The ship's soul? The ship's self? The ship's substance?
Hume’s basic psychological point was that the supposed identity of this ship is actually a result of the human “imagination” projecting a (fixed) identity onto it. Indeed, such psychological points underpin Hume’s empiricism when it comes to this subject, and to many others.
Hume’s psychological empiricism is explicitly stated on the next page when he told his readers that
“[i]dentity depends on the relations of ideas; and these relations produce identity, by means of that easy transition they occasion”.
In simple terms, identity has nothing to do with the object (say, a ship or river). Any identity we may posit has to do with the mind and its relations (to the object?). Or, in Hume’s terms, it’s to do with the relations of ideas. However, even if we relate ideas to ideas, it’s still not clear (here at least) how that would bring about the resulting identity (or the belief in an identity).
Hume then jumped from the ship of Theseus to Heraclitus’s river. (Again, without mentioning Heraclitus.) Basically, the following account is the same as the one Hume offered his readers three paragraphs earlier. He wrote:
“Thus as the nature of a river consists in the motion and change of parts, though in less than four and twenty hours these be totally altered, this hinders not the river from continuing the same during several ages.”
There are many differences between a river and a ship. However, in this context at least, they may not matter too much.
It’s hard to think of a river having “parts”. Then again, why not? There are contenders for what constitutes a river’s “identity”, such as the terrain the water moves through, etc. Yet they won’t help because Hume’s argument can be applied to these too.
Hume and Unrestricted Mereological Universalism
Hume pre-empted the position of unrestricted mereological universalism in which (to put it one way) an object can have any parts in any order. Of course, Hume used his thought experiment to advance his case against the self.
Firstly, Hume wrote: “[S]uppose any mass of matter, of which the parts are contiguous and connected, to be placed before us.” He then told his readers that
“it is plain we must attribute a perfect identity to this mass, provided all the parts continue uninterruptedly and invariably the same, whatever motion or change of place we may observe either in the whole or in any of the parts”.
Here we can say that continuity and connectedness are literally constituting the identity of this mass of matter. (That’s if we need to use the word “identity” at all.) This seems to chime in with David K. Lewis’s position, who, in his Parts of Classes (1991), wrote:
“I say that composition is unrestricted: any old things have a fusion, however scattered and miscellaneous they may be. The fusion is that one thing that overlaps all and only those things that are among the old things.”
This position isn’t a commitment to the importance of contiguity and connectedness when it comes to objects: it’s a commitment to arbitrary objects.
Hume’s final words on this subject:
“[S]uppose some very *small* or *inconsiderable* part to be added to the mass, or subtracted from it: though this absolutely destroys the identity of the whole, strictly speaking, yet as we seldom think so accurately, we scruple not to pronounce a mass of matter the same, where we find so trivial an alteration.”
This leaves us with a problem. According to Hume, adding or subtracting parts to — or from — this mass of matter absolutely destroys the identity of the whole. Yet Hume argued that we (or his philosophical opponents) believe that such a “trivial an alteration” doesn’t actually constitute a different object or mass. We deem it to still be “the same” thing.
Now let’s apply that thought experiment to the self, as Hume himself did.
From the Ship of Theseus to the Self
One of Hume’s basic points was that our view of the self as fixed (or static) is accounted for by our not noticing the big changes over time which occur to our selves. As Hume put it:
“A change in any considerable part of a body destroys its identity; but it is remarkable that where the change is produced *gradually* and *insensibly* we are less apt to ascribe to it the same effect.”
Of course, Hume was talking about a physical body here. Yet the whole point of him discussing such a thing was to get his point across about the self (or about personal identity).
Thus, I have nothing much in common with my 12-year-old self. Yet I still deem this 12-year-old self to be me. Of course, there’s a big difference between a 12 year old and a grown adult. However, all the small changes which led to the grown adult were gradual and unnoticeable. Yet that doesn’t stop it from being the case that there’s a big difference between the adult’s self and the child’s self. Thus, we may have selves in the plural when it comes to the same subject or human person.
