Is Humean Causal Skepticism Self-Refuting and or Unsound?
Hume argues that the concept of causation is attributable to the constant conjunction of events. If event x always precedes event y, we form the notion that x causes y. And since various sorts of events are constantly conjoined in our perceptions, we form various notions of causal relations, and from these various notions, we form the general notion of causality. According to Hume, we have no evidence to justify the notion that y necessarily follows from x, or that a given kind effect necessarily requires a given kind of cause, or that any effect requires any cause whatsoever. All we can know is the temporal order of our perceptions. We cannot know whether there are extra-mental relations among the objects of our perceptions such that prior states of objects determine their subsequent states. If my summary of Hume's causal skepticism is inadequate, please correct and or expand upon it.
Here is why I think Hume's account of causality is self-refuting: it presupposes a causal relation between perception and belief. Because, according to Hume, we perceive certain events constantly conjoined, we believe that they are causally related. The perception causes the belief. Hume even admits that belief in causality is very convenient, since it helps us make sense of experience. But this also entails a causal relation, since if the belief in causality helps us make sense of experience, then the belief causes us to make better sense of experience.
Now maybe Humean causal skepticism doesn't require one to self-refutingly assert that constant conjunction causes the unfounded notion of causality. Perhaps instead, the causal skeptic could simply assert that we can't prove the reality of causal necessity, and he could leave it unexplained why we think in terms of causal relations. But without constant conjunction as an explanation for causality, the causal skeptic is no longer able to answer important questions. The causal skeptic can use constant conjunction to explain, for instance, why we be believe that fire causes smoke whereas water does not cause smoke. Without constant conjunction or some other self-refuting notion, it seems that causal skepticism would undermine the explanatory power of any belief system that included it. For if I were a causal skeptic who rejected all causal explanations of causal beliefs, I could not explain why we believe that anything causes anything. I could barely answer any question that began with the word 'why'.
Maybe, I couldn't answer any questions at all, since all reasonable beliefs require justification, and all methods of justification involve some form of causality. For instance, my knowledge that the proposition 'all bachelors are married' is true is caused by my knowledge that the definition of the predicate is included within the definition of the subject.
For these reasons, causal skepticism seems unsound either way. If causal skepticism appeals to constant conjunction, it is self-refuting. If it does not appeal to constant conjunction, it makes all justification impossible, or at least all justification of causal beliefs.
Here is why I think Hume's account of causality is self-refuting: it presupposes a causal relation between perception and belief. Because, according to Hume, we perceive certain events constantly conjoined, we believe that they are causally related. The perception causes the belief. Hume even admits that belief in causality is very convenient, since it helps us make sense of experience. But this also entails a causal relation, since if the belief in causality helps us make sense of experience, then the belief causes us to make better sense of experience.
Now maybe Humean causal skepticism doesn't require one to self-refutingly assert that constant conjunction causes the unfounded notion of causality. Perhaps instead, the causal skeptic could simply assert that we can't prove the reality of causal necessity, and he could leave it unexplained why we think in terms of causal relations. But without constant conjunction as an explanation for causality, the causal skeptic is no longer able to answer important questions. The causal skeptic can use constant conjunction to explain, for instance, why we be believe that fire causes smoke whereas water does not cause smoke. Without constant conjunction or some other self-refuting notion, it seems that causal skepticism would undermine the explanatory power of any belief system that included it. For if I were a causal skeptic who rejected all causal explanations of causal beliefs, I could not explain why we believe that anything causes anything. I could barely answer any question that began with the word 'why'.
Maybe, I couldn't answer any questions at all, since all reasonable beliefs require justification, and all methods of justification involve some form of causality. For instance, my knowledge that the proposition 'all bachelors are married' is true is caused by my knowledge that the definition of the predicate is included within the definition of the subject.
For these reasons, causal skepticism seems unsound either way. If causal skepticism appeals to constant conjunction, it is self-refuting. If it does not appeal to constant conjunction, it makes all justification impossible, or at least all justification of causal beliefs.
Comments (43)
Quoting Noisy Calf
All that would prove is that Hume was inconsistent and could not follow his scepticism to the end. If that's your point, then I think you are right.
Bertrand Russell gave a very similar criticism, as follows:
That's exactly what I'm trying to get at. Very well put.
[quote=IEP;https://iep.utm.edu/kantmeta/#H2] Kant argues that the understanding must provide the concepts, which are rules for identifying what is common or universal in different representations.(A 106) He says, “without sensibility no object would be given to us; and without understanding no object would be thought. Thoughts without content are empty; intuitions without concepts are blind.” (B 75) Locke’s mistake was believing that our sensible apprehensions of objects are thinkable and reveal the properties of the objects themselves. In the Analytic of Concepts section of the Critique, Kant argues that in order to think about the input from sensibility, sensations must conform to the conceptual structure that the mind has available to it. By applying concepts, the understanding takes the particulars that are given in sensation and identifies what is common and general about them. A concept of “shelter” for instance, allows me to identify what is common in particular representations of a house, a tent, and a cave.
The empiricist might object at this point by insisting that such concepts do arise from experience, raising questions about Kant’s claim that the mind brings an a priori conceptual structure to the world. Indeed, concepts like “shelter” do arise partly from experience. But Kant raises a more fundamental issue. An empirical derivation is not sufficient to explain all of our concepts. As we have seen, Hume argued, and Kant accepts, that we cannot empirically derive our concepts of causation, substance, self, identity, and so forth. The problem that Kant points out is that a Humean association of ideas already presupposes that we can conceive of identical, persistent objects that have regular, predictable, causal behavior. And being able to conceive of objects in this rich sense presupposes that the mind makes severala priori contributions. I must be able to separate the objects from each other in my sensations, and from my sensations of myself. I must be able to attribute properties to the objects. I must be able to conceive of an external world with its own course of events that is separate from the stream of perceptions in my consciousness. These components of experience cannot be found in experience because they constitute it. The mind’s a priori conceptual contribution to experience can be enumerated by a special set of concepts that make all other empirical concepts and judgments possible. These concepts cannot be experienced directly; they are only manifest as the form which particular judgments of objects take.[/quote]
In other words, I take it, if Hume was correct, we couldn't actually think or make judgements; we'd have, at best, the cognitive capacity of a newborn, or someone with an acute neural disorder, because of the role of the mind in synthesising and constructing experience and sensations into a meaningful whole within which judgement is possible. And we don't naturally see that, because we see with it, or through it. That's why Kant can claim that in some basic sense, Hume must assume some of the very things that he purports to doubt. That's my gloss.
That makes sense. If the concept of causality is necessary to intelligibly structure experience, then all experience must presuppose it in some way. And all experience includes the experience of formulating a philosophical theory, like Hume's, that attempts to explain away causal relations.
I think Kant's notion of causality might also be self-refuting, but I'll save that for a future post.
Well, all intelligible structures of experience. I suggest that A cat can wait by a mouse hole for a mouse by means of a structured experience that we can articulate intelligibly, but a the cat cannot. My take on Hume is that his concern is to draw the limits of logic and verbal reasoning. There is a widespread materialist bias that finds the idea that one cannot derive an 'ought' from an 'is' totally convincing and irrefutable, but the idea that one cannot derive a 'will be' from a 'has been' somehow contradictory.
And in neither case is Hume making an attack on the world, that there is no such thing as causation, or that there is no such thing as morality. Rather, he is making an attack on overblown rationalism that thinks it can make the world conform to thought, instead of conforming thought to the world.
[quote=Hume]Reason is, and ought only to be the slave of the passions, and can never pretend to any other office than to serve and obey them.[/quote]
This "passion" is the rock on which Humean philosophy stands, roughly translated as 'giving a damn' or 'caring'. Reason seeks to be dispassionate, but there is no reason to reason if one does not give a damn, thus passion is what is necessary to intelligibly structure experience: the cat cares about mice, and that is the primary structure of its experience prior to any reasoning it might or might not be capable of.
Notwithstanding his explanation of why we believe in causation, 1 is either true or false and 2 is either true or false, and both being true is not a contradiction.
Hume believed that all knowledge can be categorized into "relations of ideas" and "matters of fact." On the latter he took an austere empiricist stance: matters of fact can only be known from direct sense impressions. And he argued that causation is not impressed onto us by our senses, and therefore is not a matter of fact. Neither can it be inferred by pure reason. However, he did not venture a causal theory empirical knowledge; he took the above principles as more-or-less self-evident. It is fair to dissect and question Hume's empiricist principles. It is also fair to ask whether those principles are more foundational than our causal intuitions. However, I don't think it is fair in this case to accuse him of illicitly helping himself to the very ideas that he questioned. That Hume imagined impressions being caused by their objects is a speculation.
Quoting unenlightened
That's one way to read him, but my impression is that he is somewhat less clear and more conflicted on this point. I'll leave this to Hume scholars though.
Quoting SophistiCat
What do you make of the rest of what Russell says then?:
[quote=Russell]I doubt whether either he or his disciples were ever clearly aware of this problem as to impressions. Obviously, on his view, an "impression" would have to be defined by some intrinsic character distinguishing it from an "idea," since it could not be defined causally. He could not therefore argue that impressions give knowledge of things external to ourselves, as had been done by Locke, and, in a modified form, by Berkeley. He should, therefore, have believed himself shut up in a solipsistic world, and ignorant of everything except his own mental states and their relations.[/quote]
Here Russell elaborates further on his view (the format here got a little messed up when I copied the text, sometimes that happens when using the quote function for whatever reason, so bear with it):
Also, regardless of how one views Hume's definition of an impression, it is clear that the law of habit is a causal law, and therefore Hume had no right to use it as an explanation for how the ideas of cause and effect come about, except as a sort of “reductio ad absurdum”, showing where seemingly good reasoning with sound principles has led him.
Habit is not a law at all. It is my habit to drink coffee for breakfast. But habit is not the cause of my drinking coffee, it is the mere fact that I do. Sometimes I might I have tea instead, and no law is broken, only my habit.
Do you disagree with any of that?
By the way, I said “the law of habit” not “habit”, they are not quite the same thing.
Quoting unenlightened
That’s not what Russell is saying, he’s saying that my expectation that when I see an apple in the future, it will taste like how apples usually taste and not like roast beef, is explained by the fact that I have always expected apples to taste that way in the past whenever I see them.
Thus, although in the past the sight of an apple (cause) has been conjoined with expectation of a certain kind of taste (effect), that gives no justification for the claim that it will continue to be so conjoined in the future.
Basically, expectations also fall prey to the problem of induction.
However in the case of expecting causation and the world to go on being predictable in the future, the habit is always there. So it's not like usually drinking coffee in the morning. It's expecting that there will be such a thing as coffee to drink, which will have a certain flavor and caffein content that has some stimulating effect on your nervous system.
Yes, the expectation is not justified. That's why it's just a habit and not a reason. That there is no rational justification is what Hume is saying. We expect the future to be like the past because NO REASON, we just do it habitually. What is daft is to claim that Hume needs to justify his habits when he's just said there is no justification for them.
And there is no justification for it. That is what the man says. We do it, and reason cannot justify it.
No wonder Kant was worried about science after Hume. If we take Hume seriously, there is no reason for anything that happens. Therefore, there are no explanations. Just descriptions of constant conjunction, up to this point. Science is a fancy kind of book keeping.
My constantly conjoined habit results in me wondering why there is such a vast constant conjunction of events throughout the observable universe. It being radically contingent like that beggars belief. So does the thought that it could change at any moment, for no reason.
My habit means I can't buy Humean skepticism.
I disagree, since theories need to also be predictive. That and every scientist talks in terms of explanation and cause when they're not waxing philosophical.
Quoting unenlightened
It's not just that we may be wrong about how future apples will taste, but also that we have no reason to expect that in the future we will even expect that they will have their usual taste instead of expecting that they will taste like ice cream.
[quote=Russell] I do not wish, at the moment, to discuss induction, which is a large and difficult subject; for the moment, I am content to observe that, if the first half of Hume's doctrine is admitted, the rejection of induction makes all expectation as to the future irrational, even the expectation that we shall continue to feel expectations. I do not mean merely that our expectations may be mistaken; that, in any case, must be admitted. I mean that, taking even our firmest expectations, such as that the sun will rise
tomorrow, there is not a shadow of a reason for supposing them more likely to be verified than not.[/quote]
That is to say: if Hume is right, then not only is our expectation that the sun will rise tomorrow not justified, but neither is the expectation that we will continue to expect the sun to rise tomorrow justified.
If that's the case, Hume has no right to say such things as:
Quoting Amalac
[quote=Hume]]Whoever has taken the pains to refute this total scepticism, has really disputed without an antagonist,[/quote]
Do you not see that he is talking exactly about the argument you are making? He agrees with you that he has no right to say such things, to the extent that you mean he has no justification in reason. However, everybody does think, say and presume such things, himself included.
So what you're saying is that we have no reason to think we'll have a habit of expecting things in the future. If there is no reason for constant conjunction, that includes our habits in the future.
That's what Russell is saying would be the case if Hume was right (if I understood him correctly).
As I said at the outset and @Amalac's quote above confirms, Hume is not trying to deny causation any more than he is trying to deny morality. He is showing the limits of reason. It is certainly unreasonable to demand that he exceed the limits he is trying to establish.
Any philosophical approach that utilizes a notion of refutation courts skepticism.
If that's all Hume were trying to do, then I would happily get behind him. But I'll leave the discussion of Hume's belief system to others. Unfortunately for people like me who enjoy speculation more than research, an ism that takes its name from a philosopher is probably less nuanced and compelling than the ideas of the philosopher it's named after.
Quoting Michael
I agree that those two premises are not refuted, at least not directly, by the conclusion that we have no knowledge of causality. But as you seem to admit, the conclusion does refute Hume's (apparent?) notion that we can reasonably identify constant conjunction as the source of our belief in causality and our expectations regarding future events. And in the the third and fourth paragraphs of my original post, I argued that a form of causal skepticism that doesn't offer an explanation for our causal beliefs is explanatorily inadequate, and should thus be rejected as unsound. Any belief that makes it impossible to explain key facts about reality, I think, ought to be rejected.
And maybe the argument you presented is indirectly refuted by non-explanatory causal skepticism. For if we don't have a way of explaining why we believe some things to cause other things, we can't rationally distinguish between causal and non-causal relations. Thus, we cannot justify believing in any causal relations, whether these causal relations are real and physical or merely intuitive and psychological. But the idea of knowledge seems to presuppose at least some form of causal relation. For the causal realist, the mental act of justification really causes a belief to become an item of knowledge. For the Humean skeptic, the fact that the mental act of justification always precedes beliefs becoming more firmly held suffices to differentiate, at least from a psychological standpoint, mere belief from what we call knowledge. But for the non-explanatory skeptic, who denies constant conjunction as an explanation for our causal intuitions, nothing seems to differentiate knowledge from unjustified belief. Therefore, if non-explanatory skepticism is true, we can have no reason to believe either of the two premises you presented. Can you find a flaw in this argument? If not, then perhaps causal skepticism is self-refuting whether or not it tries to explain causal beliefs, although non-explanatory causal skepticism refutes itself in a less direct way.
Quoting SophistiCat
I seem to remember Hume illicitly helping himself to causal notions, but perhaps my memory is foggy and or my reading was uncharitable. But others in this thread seem to be debating exactly what Hume meant, so I will leave that to them.
Quoting Joshs
Are you saying that philosophical skeptics always employ a kind of self-destruct mechanism that they believe is rooted within the structure of human reason? For instance, skeptics will often point to paradoxes as evidence that reason cannot offer a consistent explanation reality. Is Hume, by showing that belief in causation is unreasonable, showing that any kind of belief, including belief that belief in causation is unreasonable, is unreasonable? This reminds me of the Wittgenstein quote about the ladder which, once it gets you to the final conclusion, you must kick away. Once reason succeeds at undermining itself, you can just stop using reason, so you don't need reason to justify your skepticism. Is this what you're getting at?
It may be that there is no such thing as causation, or that causation is inexplicable, or that there is an as-of-yet unknown means to explain causation. If one of these is true then I don’t think Hume’s argument fails.
After all, I don’t need to provide an alternative to a God of the gaps to argue that a God of the gaps explanation is either false or unjustified.
In the original post and in my response to you, I differentiated Hume's causal skepticism (some in the thread have argued that Russel and I are wrong to attribute this to Hume, but so be it), which self-refutingly explains our causal beliefs by appealing to constant conjunction, and non-explanatory causal skepticism, which doesn't attempt to explain our causal beliefs at all. And I made two arguments 1) non-explanatory causal skepticism fails to explain key facts about reality 2) non-explanatory causal-skepticism is specifically necessary to account for human knowledge. By key facts about reality, I just mean things like why are cheetahs faster than dogs, or why do people slip on wet floors, or why is it dangerous to drive blindfolded. Without at least some vague notion of causality, I don't think you could answer any of these questions.
As for 2), why non-explanatory causal-skepticism is specifically necessary to account for human knowledge, refer to the arguments in my previous posts. If there is no causation, I argued, there can be no knowledge. So if causation is inexplicable, then knowledge is inexplicable, and we can't explain the difference between justified knowledge and unjustified belief. If there is an as-of-yet unknown means to explain causation, then the means of explaining what makes some beliefs justified and others not justified is also as-of-yet unknown, so we cannot know the difference between knowledge and unjustified belief. All three of these conclusions would invalidate all of our knowledge and lead to radical skepticism. For if we can't know anything, we have no justification to belief anything. If radical skepticism is true, then philosophy is futile. So I think radical skepticism is also self-refuting, since it can't justify itself.
Quoting Michael
I think that depends on what you mean by God of the gaps. I think there are valid and invalid gap arguments. An invalid gap argument would go like this: 1) P is a fact that we have no explanation for, 2) if Q were true, Q would explain P, 3) therefore, Q is true. But a valid gap argument would be: 1) either P is true or Q is necessarily inexplicable, 2) If Q is necessarily inexplicable, then it is impossible to provide a coherent account reality, 3) therefore P is true. My argument is of the latter kind, not of the former kind: 1) Either it can be explained why certain things are causally related, or human knowledge is necessarily inexplicable, 2) if human knowledge is necessarily inexplicable, then it is impossible to provide a coherent account of reality, and radical skepticism ensues, 3) therefore, it can be explained why certain things are causally related, namely, why the mental act of justification causes belief to become knowledge.
This is not true, I don't think so. The law of habit is not a law of causation. It is simply a recurring occurrence, much like every other law has recurring occurrences that support that particular law. The fact that people see laws where there are no laws, can be due to a recurring coincidence or people seeing laws where there are none. There is no self-contradiction there.
I know the consensus is there, but in my humble opinion, the consensus is wrong.
Hume's opinion is not a refutation of the universe of causation and determinism; it is a parallel explanation to it. You don't need to believe it, but you have to accept, that it's a valid way of looking at things.
Hume's world of coincidences, and the world viewed as a series of causations, are both valid, but mutually exclusive.
At the risk of making you lose your patience again... isn’t this relation between the sight of an apple and the expectation of a certain kind of taste, an instance of the causal law of habit?:
Quoting Amalac
We expect that we will expect apples to taste how they usually taste, because we are habituated to think that way, and so we expect that we won’t cease to expect them to taste like they usually taste, and not like ice cream in, say, the next 5 minutes. But even the truth of this claim depends on the validity of induction from particular instances, and so it is not even founded on probability:
[quote=Russell]if we take Hume seriously we must say: Although in the past the sight of an apple has been conjoined with expectation of a certain kind of taste, there is no reason why it should continue to be so conjoined: perhaps the next time I see an apple I shall expect it to taste like roast beef. You may, at the moment, think this unlikely; but that is no reason for expecting that you will think it unlikely five minutes hence. If Hume's objective doctrine is right, we have no better reason for expectations in psychology than in the physical world.[/quote]
our expectation has nothing to do with laws. Our laws are based on fulfilled expectations, but our expectation in and by itself does not a law make.
Hume would also probably say that our expectations are coincidental, too. Tomorrow you may wake up and expect apples to taste like watermelons.
I'm not sure about that, it seems that if one accepts that the problem of induction also applies to psychology, Hume has no right to say things like these:
Because what he says about “Nature” either assumes the validity of induction when applied to expectations of similar expectations for the future, or is not justified by any reason, having neither certainty nor probability (that is: no reason to think it more likely than its negation), and therefore there is no reason to believe it.
Basically, we can't be sure that our nature will remain constant so as to always make us expect reasonable things. There is no rational justification even for the claim that I will probably not expect apples to taste like watermelons tomorrow.
Quoting god must be atheist
I think you can work it out from here.
Yes, but there are some difficulties for Humean causation. For one, it can't differentiate between a state of affairs which is said to be impossible, and one that just never happens. Take the examples of a perpetual motion machine or accelerating up to the speed of light. Both are ruled out as impossible by physics. Now take a river of soda. There's nothing impossible about it, but probably it will never happen because who is going to bother to make such a thing?
A second difficulty is that it renders all human behavior as coincidental. There's no reason for any of our actions. So why hold people accountable? Why blame them for anything they do?
A third issue is that it collapses the distinction between determinism and indeterminism. We can't say that the wave function evolves deterministically because prior states are conjoined for no reason.
Quoting Marchesk
The problem of induction goes even further actually: although in the past the laws of physics have not changed, that doesn't justifiy the expectation that they won't change in the future.
Quoting Marchesk
Let me turn that around for you: Why should we not hold people accountable? Why should we not blame them for the bad things they do?
I'm playing devil's (Hume's) advocate here.
True, so perpetual motion machines and FTL acceleration drives could become possible in the future if there's no reason for the constant conjunction to continue.
Quoting Amalac
Because there's no reason for them doing bad things. Same for good things as well. Blame and praise, punishment and reward are pointless. Of course we'll continue to do those because of habit, or will we???
If punishing an evil person is pointless, then choosing not to punish them is just as pointless.
If praising someone who is good is pointless, not praising them is just as pointless.
Quoting Marchesk
That's what I was hinting at, and yes: We can't do anything except believe that, even though we have no reason for doing so (but then again, we don't have any reason not to believe it either).
Still, it's pretty hopeless if Hume is right.
Russell seems to assume an externalist, causal conception of empirical knowledge, and then projects that assumption onto Locke and then Hume, for whom this would be a stolen concept, given his stance on causation. But whether or not Russell is right about knowledge, the question here is whether Hume espoused the same view: only then would he be open to the charge in the OP. Hume actually seems to hold an internalist view, at least some of the time, i.e. his account of knowledge refers only to mental states. But Russell may well be right in that looking carefully through his writing, one can catch him out on a contradiction.
I don't think he claims this to be the logical conclusion of his argument. It is more like, having cleared the ground, he is venturing a plausible hypothesis. As any narrative that describes events and processes, it must rely on causal notions in order to make any sense. But he already admitted that causality is ingrained in our thought process and is indispensable for sense-making.
You raised some excellent points.
1. Impossible vs something never occurring. A thing that never occurs is not a matter of interest. Whether because it's impossible, or just highly improbable.
3. The third issue is already spoken for -- if I understand it clearly -- which is coincidentiality. If a wave form or whatever occurs, it does, not becuase it is caused, either by determinism, or indeterminism. It is of no concern to the Humean world view.
2. Your second point is the most interesting one. I quoted it in the beginning of my post here. The accountability for actions is not there, but it's equally not there in a causational world. A person who commits a crime or other unwanted action, is determined to do so by the causing effects of past events. It is not fair therefore to punish them. Much like in Hume's world. You can't blame anyone for wrongdoing in the causational world, either.
Quoting SophistiCat
I think this passage of his Enquiry does seem to show that Hume did not hold an externalist conception of empirical knowledge:
[quote=Hume]It seems also evident, that, when men follow this blind and powerful instinct of nature, they always suppose the very images, presented by the senses, to be the external objects, and never entertain any suspicion, that the one are nothing but representations of the other. This very table, which we see white, and which we feel hard, is believed to exist, independent of our perception, and to be something external to our mind, which perceives it. Our presence bestows not being on it: our absence does not annihilate it. It preserves its existence uniform and entire, independent of the situation of intelligent beings, who perceive or contemplate it.
But this universal and primary opinion of all men is soon destroyed by the slightest philosophy, which teaches us, that nothing can ever be present to the mind but an image or perception, and that the senses are only the inlets, through which these images are conveyed, without being able to produce any immediate intercourse between the mind and the object. The table, which we see, seems to diminish, as we remove farther from it: but the real table, which exists independent of us, suffers no alteration: it was, therefore, nothing but its image, which was present to the mind. These are the obvious dictates of reason; and no man, who reflects, ever doubted, that the existences, which we consider, when we say, this house and that tree, are nothing but perceptions in the mind, and fleeting copies or representations of other existences, which remain uniform and independent.
119. So far, then, are we necessitated by reasoning to contradict or depart from the primary instincts of nature, and to embrace a new system with regard to the evidence of our senses. But here philosophy finds herself extremely embarrassed, when she would justify this new system, and obviate the cavils and objections of the sceptics. She can no longer plead the infallible and irresistible instinct of nature: for that led us to a quite different system, which is acknowledged fallible and even erroneous. And to justify this pretended philosophical system, by a chain of clear and convincing argument, or even any appearance of argument, exceeds the power of all human capacity.
By what argument can it be proved, that the perceptions of the mind must be caused by external objects, entirely different from them, though resembling them (if that be possible) and could not arise either from the energy of the mind itself, or from the suggestion of some invisible and unknown spirit, or from some other cause still more unknown to us? It is acknowledged, that, in fact, many of these perceptions arise not from anything external, as in dreams, madness, and other diseases. And nothing can be more inexplicable than the manner, in which body should so operate upon mind as ever to convey an image of itself to a substance, supposed of so different, and even contrary a nature.
It is a question of fact, whether the perceptions of the senses be produced by external objects, resembling them: how shall this question be determined? By experience surely; as all other questions of a like nature. But here experience is, and must be entirely silent. The mind has never anything present to it but the perceptions, and cannot possibly reach any experience of their connexion with objects. The supposition of such a connexion is, therefore, without any foundation in reasoning.[/quote]
By the same token, Hume could admit that he cannot prove that impressions are caused by external objects.
It seems to me Russell's point is that Locke's thesis that all ideas are copied from impressions is implausible if that's the case. That seems kind of unreasonable though: Russell is basically saying that Hume cannot define impressions like that unless he can refute external world scepticism.
Hume admits he cannot do that, and that neither can anyone else (including Russell). Yet that doesn't stop Russell from developing philosophical theories which are only plausible if there is an external world.
So yes, I think you might be right after all.
Quoting Goodman: Fact Fiction and Forecast, p61
I suppose it makes sense that constant-conjunction is only a hypothesis, although I haven't read Hume recently enough to judge whether or not he phrased it as a mere hypothesis. However, the hypothesis seems to entail that causality is real, and if causality is real in psychology, it might as well be real in physics and other realms also. But I guess the point of the constant-conjunction hypothesis might not be to reject the reality of causality. Rather, the point might be to show that, given the inductive manner whereby we arrive at the notion of causality plus the problem of induction, we cannot have certain knowledge regarding the reality of causal relations. But as I've argued elsewhere in this post, I think causal relations are indispensable to the justification of belief and hence to the existence of knowledge. So skepticism regarding causal relations, I've argued, entails skepticism regarding all knowledge i.e. radical skepticism. And I think radical skepticism is a self-defeating position, since if knowledge is probably impossible, then one probably cannot know whether or not knowledge is possible, in which case one is not justified in believing that knowledge is probably impossible. If my arguments are sound, then causal skepticism can be refuted by retorsion.