For Hume, ‘perceptions’ form the entirety of the content of one’s mind. Perceptions can be further divided into impressions and ideas. Impressions are all the experiences of the five senses and our passions and desires, as originally felt. They enter the consciousness with ‘most force and violence.’ Ideas are the memories or re-awakenings of past impressions and are weaker and less forceful. Another division can be made between simple and complex perceptions. Simple impressions and ideas admit of no distinction or separation, they can almost be compared to the atoms of experience. Complex perceptions are closer to molecules; complex perceptions are composed of simple perceptions, and can hence be analysed in terms of these simple perceptions. Hume allows that one can have complex ideas that are derived from no corresponding impression, but that all simple ideas must be derived from corresponding simple impressions. Furthermore, each simple impression must differ only in degree of force and vivacity, such that there is a direct resemblance between simple impressions and simple ideas.  

Hume argued that the meaning of a given term is the idea associated with it, and hence, if a philosophical term lacks an associated idea, then it similarly lacks meaning. Since meaningful terms must have associative ideas, and hence must ultimately derive from experiences, we are offered a form of meaning-empiricism: terms must be based on experience in order to have meaning. Hence, meaningful terms must have corresponding impressions. Hume admits that his ‘Copy Theory,’ whereby all ideas must derive from impressions, allows for exceptions. He gives ‘the missing shade of blue’ example, wherein, an individual is given a gradated chart of all the shades of blue. Although he has never experienced one particular shade of blue, he determines an idea of that shade when looking at a gap in the colour chart. The question is whether this has any bearing on issues regarding philosophical terms. The question to ask is this: ‘if the man then applies a term to the idea of this particular shade of blue, will this term be meaningful just because he has a corresponding idea, or will it be meaningful because the idea can be experienced as an impression?’ If the former is the answer, then any idea, be it ‘soul’ or ‘substance,’ or what have you, would be rendered meaningful for the same reason that the idea of the shade of blue is, a consequence that Hume would not want to accept. If the latter is the answer, then his idea is meaningful only because the shade of blue could be experienced as an impression. Here we have circularity – we can only know whether something can be experienced as an impression once we have had that impression. 

If we use the term ‘idea’ in the strict Humean sense of being (the only) re-imagined simple impressions or complexes of different re-imagined simple impressions (which I will from now on refer to as mental copies), then we must conclude that, under Hume’s own definitions, a term without any corresponding idea will necessarily be derived from no impressions, and vice versa. This is not the contention; the contention is that Hume equates the meaning of a word with a mental copy. The self may, under Hume’s definition, not be an idea (since the self has no corresponding mental copy) but the meaning of a word simply isn’t a mental copy, it is closer to a concept. Although there may be no corresponding impression for the term ‘self,’ is not meaningless for a number of reasons: first, we can use the term, second we can apply predicates to the term, and third, we have a concept of the term. One may also invent meaningful terms that by their very definition lack any corresponding impression. For example, ‘X’ could refer to ‘any object of which it is impossible to have an impression.’ Here, we have a meaningful term which not only lacks a corresponding idea, but necessarily lacks a corresponding idea.  

The plausibility of Hume’s account depends partly upon whether he regards terms as being meaningless on the sole grounds of having no corresponding impressions, or whether he takes it to be a contingent truth that all meaningful terms have thus far been derived from impressions. We must note that Hume has no apriori reason for supposing that all meaningful philosophical terms must derive from experience. Resultantly, Hume’s reasoning must be based on aposteriori or experiential reasoning. If, however, his only reason for suggesting that terms can lack meaning is because they lack corresponding impressions, then Hume will be extrapolating from ‘all terms that thus far have lacked corresponding impressions have similarly lacked meaning,’ to ‘any term that lacks a corresponding impression will similarly lacks meaning.’ Hume can then reject any claim to a meaningful term on the basis that it lacks a corresponding impression – this here will be an example of Hume taking a contingent generalisation and turning it in a necessary truth. 

Some may argue, however, that Hume does not dismiss philosophical terms as meaningless on the sole basis of lacking corresponding impressions, but rather, we already have suspicions of meaninglessness, and that, on observing that there is also no corresponding impression with the term, our suspicion is confirmed. Yet, Hume gives no other account for why a term might be meaningless. The lack of impressions is not being provided as an additional piece of evidence among many, rather, it is the piece of evidence responsible for confirming our suspicions, and it is the thing that makes it more than a suspicion. I therefore agree with Flew’s estimation that Hume makes the error of turning a contingent fact into a necessary truth. 

Nonetheless, I would argue that the contingent fact itself is largely true. Almost all meaningful terms must be derived from impressions, at least indirectly (otherwise no thing will be represented by the term). Forces and the names of sub-atomic particles are invariably meaningful despite having no corresponding impression. Even so, these things are still inferred from other impressions. So although subatomic particles are not derived directly from impressions, they are still derived indirectly from impressions. My criticisms of Hume are two: first, he regards the meaning of a term to being synonymous with a mental copy, second; he extrapolates the general rule that all meaningful terms are derived from impressions into a necessary truth that all meaningful terms are derived from impressions.