I’ve been thinking recently about Berkeley’s views on language. I just taught the Three Dialogues Between Hylas and Philonous last term, and felt like Berkeley was pulling some philosophical sleight of hand in Philonous’s reply to Hylas’s parity argument against spirit (in dialogue three). Regardless of what is going on with the challenge to material substance and Hylas’s parity argument against spirit, it is clear that Berkeley thinks we do not have an idea of spirit. And yet, Berkeley holds that the term “spirit” is significant, and does not appear to think we should avoid using it. One place to look for some illumination on how Berkeley thinks this could be comes from the dialogues “Alciphron, or the minute Philosopher”.
In Alciphron VII, Berkeley has Euphranor explains that we should “see if we an make sense of our daily practice” in using words as signs, and so says to Alciphron:
Words, it is agreed, are signs: It may not, therefore, be amiss to examine the use of other signs, in order to know that of words. Counters, for instance at a card-table are used, not for their own sake, but only as signs substituted for money, as words are for ideas. Say now, Alciphron, is it necessary every time these counters are used, throughout the whole progress of a game, to frame an idea of the distinct sum or value that each represents?
Alciphron concedes that poker chips work fine as signs, as long as there is a settled agreement as to how to turn the counters back into money at the end of the game. Euphranor then pushes the point, asking whether we need to think about pounds, shillings and pence when we are tabulating a sum. Alciphron agrees that the important thing is whether, “in the conclusion, those figures direct our actions with respect to things”. This leads Euphranor to conclude:
From Hence, it seems to follow, that words may not be insignificant, although they should not every time they are used, excite the ideas they signify in our minds, it being sufficient, that we have it in our power to substitute things or ideas for their signs when there is occasion. It seems also to follow, that there may be another use of words, besides that of marking and suggesting distinct ideas, to wit, the influencing our conduct and actions; which may be done, either by forming rules for us to act by, or by raising certain passions, dispositions, and emotions in our minds. A discourse, therefore, that directs how to act, or excite to the doing or forbearance of an action may, it seems, be useful and significant, although the words whereof it is composed, should not bring each a distinct idea into our minds. (emphasis mine)
What I find fascinating about this exchange is the degree to which it rejects the (broadly Lockean) cognitivist paradigm of language.
What I mean by “cognitivist” concerns the relationship between language and mind. The broadly Lockean paradigm is cognitivist insofar as the central account of the meaningfulness of terms is given in terms of ideas in the understanding, and the central account of the significance of sentences is given in terms of the cognitive activity of judgment. Various features of verbal propositions are explained in terms of (and had derivative from) the features of the mental states they express, and those mental states are, by and large, from the cognitive side of things.
Berkeley, through Euphranor, is offering a pretty hefty overhaul of this picture, though he does not deny that in many cases words are used in that broadly Lockean fashion. First, Euphranor got Alciphron to concede that there are individual uses of significant terms that are not, on that occasion of use, backed by an idea in the understanding of the speaker. This is not, by itself, a deep difficulty for the cognitivist paradigm, as the cognitivist could simply account for the phenomenon as a sort of derivative use, only possible if there have been directly significant uses previously (or only if that use stands in the right relationship to directly significant uses). Nothing yet is distinctively non-cognitive, because so far, we are just talking about cases in which you use words without the ideas that they stand for. Since the words still stand for ideas, though, this is not a major break from the basic picture.
The break comes in the second of the two passages I quoted, where Euphranor suggests that the broader lesson to draw from such cases is that words have a use, other than marking and suggesting ideas: influencing our conduct and actions. Here we get a picture on which a term is significant, despite there being no corresponding idea, provided that the term influences our conduct and actions (either by being the formation of a rule of action, or by raising passions/dispositions/emotions to mind). This is precisely the sort of view of the workings of language that one needs in place for some of the traditional versions of non-cognitivism about ethics. On those views, the word “wrong” is meaningful because its use is connected with the non-cognitive state of disapproval (in some way).
Some of Berkeley’s examples of terms which require this distinctive apparatus are: self, number, force, grace, trinity, substance, personality. Further, Berkeley (through Euphranor) offers a non-cognitivist account of the doctrine of the trinity, saying:
[A] man may believe the doctrine of the trinity, if he finds it revealed in Holy Scripture, that the Father, the Son, and the Holy Ghost are God, and that there is but one God? Although he doth not frame in his mind, any abstract, or distinct ideas of trinity, substance, or personality, provided, that this doctrine of a Creator, Redeemer, and Sanctifier makes proper impressions on his mind, producing therein, love, hope, gratitude and obedience, and thereby becomes a lively operative principle, influencing his conduct and actions, agreeably to that notion of a saving faith which is required in a christian. (p. 348 of the text linked above).
I’ve become pretty fascinated by this thread of thinking in Berkeley recently, though I haven’t had a chance to start digging into the secondary literature on it yet. What I am most interested in determining is whether or not Berkeley can address the concern that non-cognitivism requires anti- or quasi- realism about the domain in question.
When Locke defines truth for a verbal proposition, he does so in a way that makes it dependent on the mental proposition expressed. This model, on which linguistic activity inherits many of its interesting features from mental activity, produces a relatively short argument for the non-truth-evaluability of verbal propositions which receive a non-cognitivist treatment. If a sentence/utterance/assertion can only be true or false insofar as the mental state that it expresses is true or false, then a sentence which expresses a non-cognitive state like desire would not be true or false. I take it that this conclusion is anti-realist about the domain in question. One approach that is popular as of late is to go minimalist about truth, a maneuver which yields quasi-realism about the domain in question. I take it though, that there is good reason to suspect that Berkeley is not a mere quasi-realist about the existence of God or the self. His non-cognitivisms appear to be bred out of facts about limitations of mental representation, and not out of a suspicion about the reality of God or the self. It would be nice figure out if there is some way to capture this difference between the sorts of non-cognitivist views which are motivated by (or coupled with) a rejection of the metaphysical reality of the categories in question, and those which seem to, at the end of the day, want to say something more like, “oh, they’re real all right, we just aren’t equipped to think about those things”.
Lewis Powell
Rico Vitz
beckocopenhaver
LisaShapiro
lebuffe
Antonia LoLordo
juliekrwalsh
Kenny Pearce
lizgoodnick
Stewart Duncan
Iulia Mihai
Colin Heydt
kateamoran
Joshua M. Wood
Eugene Marshall
samrickless
Sydney Penner
Marcus Adams
jennifermarusic
Colin McLear
Justin Steinberg
marcy p lascano
kristenirwin
What a great post. Have you looked at the Querist yet? I think it provides quite a bit of grist for your non-cognitivist mill….
The Querist is the one that is written entirely in interrogatives? I think I’ve looked at it briefly, but I definitely haven’t given it a careful going-over.
[...] more radical rejection of the necessity of ideational signification for meaningful terms (as I have discussed before), it is quite interesting that Leibniz, Berkeley, and Hume all seems to recognize some potential [...]
Lewis, this is an interesting post. I don’t know the literature either, but this won’t stop me from asking a quick question here.
I seems like we can distinguish between two forms of cognitivism about the term ‘spirit’: (1) this term is used meaningfully just in case the mental state to which it refers is presently “excited in the mind”; and (2) this term is used meaningfully just in case it is possible for the mental state to which it refers to be “excited in the mind” (where such possibility stems from prior acquaintance with the relevant mental state).
The passages you cite strongly suggest that Berkeley denies (1). But it is unclear to me whether Berkeley denies (2). My assumption is that if Berkeley is a non-cognitivist about rich philosophical terms like ‘spirit,’ ‘self,’ and ‘God,’ then he must deny (2). If this assumption is right, then it is worth noting that Berkeley, in at least one passage, appears to endorse the version of cognitivism outlined in (2). Berkeley writes: “words may not be insignificant, although they should not every time they are used, excite the ideas they signify in our minds, it being sufficient, that we have it in our power to substitute things or ideas for their signs when there is occasion.”
Of course, Berkeley explicitly denies that we have an “idea” of spirit. How do you take this to fit into his non-cognitivism? In particular, do you take this to constitute Berkeley’s denial of (2)?
Just a bit of terminological clarification: I don’t think that “spirit” refers to a mental state for Berkeley, but to a mental substance. So let me spell out the generalized versions of the two views I take you to be asking about:
1′: A term T (with referent R) can be used meaningfully just in case, on the occasion of its use, the speaker has an idea of R excited before the mind.
2′: A term T (with referent R) can be used meaningfully just in case, on the occasion of its use, the speaker has a capacity to excite an idea of R.
I agree that these are two different strengths of “cognitivist” requirement, and that a genuine non-cognitivism would require that even 2′ be rejected. Additionally, I take it that Berkeley’s initial example with counters standing in for money is an example of Berkeley indicating that 1′ is too strong of a requirement, and signaling (at least) a shift to 2′.
I would say, though, regarding the passage you cite, that Berkeley is not there endorsing the biconditional version of 2′, but rather, endorsing only the direction that says, roughly, “If you use the term and have the capacity to bring the idea before your mind, then you are using it with significance.”
I actually should also say that I think one could further weaken 2′ and still be a cognitivist. Suppose that I (like Leibniz, Hume, and Berkeley) think that there are cases where we substitute manipulation of symbols for manipulation of ideas in reasoning, and especially in math. So that, ordinarily, when I am doing math or geometry, I am engaged in a sort of “blind thought”. Now, for some cases, it is clear that I satisfy (2′). I may not actually token an idea of some group that numbers 38 when calculating 38 divided by 2, but I could. It is not clear that this is true for arbitrarily large numbers (Descartes’s discussion of the chilleagon comes to mind here). It might be that a cognitivist could allow for some sort of recursive account of the meaningfulness of “1,234,777,983.21″ in virtue of that numeral being part of a system, the fundamental elements of which we can excite ideas of, cashed out in terms of the inferential connections between the unthinkable numbers and the numbers that we are able to think of.
I think I take Berkeley’s denial of two to come from the bolded line in the passage I quoted in the main post: ” It seems also to follow, that there may be another use of words, besides that of marking and suggesting distinct ideas, to wit, the influencing our conduct and actions; which may be done, either by forming rules for us to act by, or by raising certain passions, dispositions, and emotions in our minds.”
The reason I take this to be the commitment to non-cognitivism is simply that, in the traditional division of the mind’s faculties into the understanding and the will, cognitive states belong to the understanding and conative states belong to the will. An occurrent passions is not an idea, a judgment, or a bit of reasoning (at least, not the way Berkeley is classifying them), and so, is not properly part of the cognitive mind. Since words can be meaningful in virtue of exciting passions, emotions, or rules of conduct, this is quite literally a case in which the meaningfulness of some terms gets explained in a non-cognitive fashion.
Thanks, Lewis, this is really helpful.
I think you are right to claim that Berkeley denies (2), since the boldfaced passage, as you point out, suggests as much. Berkeley still subscribes to some version of cognitivism with respect to the significance of some terms. Unless I’m mistaken, I believe you grant as much. The present challenge is to show why Berkeley adopts a cognitivist explanation of the significance of some terms while opting for a non-cognitivist explanation of the significance of other terms.
As I understand matters, you appeal to the traditional distinction between will and understanding as a way to explain why Berkeley adopts a non-cognitivist explanation in the case of terms like ‘joy’ and ‘volition.’ I’m curious about the following. Does this traditional distinction itself push us in the direction of giving a non-cognitivist account of the significance of the terms ‘joy’ and ‘volition’? Or does Berkeley construe this distinction in such a way that entails the need for a non-cognitivist account about the meaning of these terms?
I have doubts about the former. Locke has a cognitivist account of the meaning of terms like ‘joy’ and ‘volition.’ I believe Hutcheson does too. Both thinkers, in accordance with the traditional distinction between will and understanding, are prepared to grant that the referents of ‘joy’ and ‘volition’ do not belong to the understanding. But they do not take this fact to entail the need for a non-cognitivist account of the meaning of these terms.
Have they made a mistake here? As it seems to me, there is nothing about the traditional distinction itself, and the fact that these terms refer to “conative” rather than “cognitive” states, that precludes a satisfactory cognitivist explanation of the significance of the terms ‘joy’ and ‘volition.’ For, at least according to Locke and Hutcheson, mental states, whether they belong to the will or to the understanding, are all accessible by means of introspection. As such, the meaning of the terms referring to conative states are leached from a mental state in the same way as the meaning of terms referring to cognitive states.
Here, as it seems to me, is where Berkeley enters. He denies that we have any “idea” of volition or any “ideas” of the passions. And I take this claim to constitute, at minimum, the denial that introspection, which can acquaint us with mental states belonging to the understanding, can also acquaint us with mental states belonging to the will. However, it is not clear to me that this denial implies that Berkeley is committed to a non-cognitivist account of the meaning of the terms ‘joy’ and ‘volition.’ His denial could imply as little as that we are acquainted with volition and the passions by some means other than introspection. Of course, we could pursue a stronger reading of this denial. It might be taken to imply that it is impossible for the meaning of the terms ‘joy’ and ‘volition’ to be leached from any mental state. This, as it seems to me, is how we must read Berkeley’s denial in order for the distinction between will and understanding to entail that he is a non-cognitivist about the terms ‘joy’ and ‘volition.’ Perhaps I am wrong about this.
In any case, I’d be interested to hear more about how you take the distinction between will and understanding to relate to Berkeley’s non-cognitivism.
Here’s how I think about it, in very broad strokes: There is a sort of generally Lockean, thoroughly Psychologistic paradigm about language that was popular in the early modern period. The fundamental tenet of this paradigm is that the philosophical treatment of language will piggyback on the philosophical treatment of mental states. So, for instance, ordinary declarative sentences are to be understood principally in terms of the beliefs they express, and only derivatively in terms of the descriptive claims they make or the conditions of their truth. Locke is quite explicit that verbal propositions are true or false in virtue of signifying mental propositions, which are the primary truth-bearers for him. This sort of picture requires that the sub-sentential bearers of meaning are given psychologistic treatment as well. This is generally spelled out as the doctrine that words signify ideas, since ideas are the relevant sub-components of judgments. Thus, the story about ordinary boring declarative sentences on this Lockean paradigm is one that assigns ideas to terms and judgments to sentences. The sentence “Snow is white” is the verbal expression of the judgment affirming the idea WHITE of the idea SNOW. The elements that make this a cognitivist picture are that complete sentences express judgments (clearly on the cognitive side of things) and meaningful parts of sentences express ideas (also clearly on the cognitive side of things).
So for Locke, the word “Joy” directly signifies the idea of joy. That is: not the passion of joy itself, but the idea of that passion. This is still cognitivist, because the sentence “Joy is wonderful” expresses the judgment affirming WONDERFULNESS of JOY.
For contrast, consider a (fairly naive) reading of Hume’s theory of moral thought: To judge something to be bad is to be actuated by the sentiment of moral disapproval upon considering it. This is a non-cognitivist treatment of “Murder is wrong” because instead of expressing the cognitive judgment affirming the idea of wrongness of the idea of murder, it expresses the (moral) disapproval produced on the contemplation of murder.
Now, talk about joy (that is, discourse in which, ultimately, we are making claims about joy) is easy to give a cognitivist treatment, so long as one has an account of how we can think about joy. But suppose you want to give an account of the exclamation “Yippee!”. Uttering “Yippee!” is not making a claim that you are experiencing joy. So if joy is involved in the story about “Yippee” it will be in a different way than it is involved in the story for “Joy is wonderful”. Being joyful is more like the sincerity condition for “Yippee” than it is part of the truth conditions (after all, “Yippee” doesn’t even have truth conditions). Just as believing that snow is white is the sincerity condition for “Snow is white”. Saying “Snow is white” is not making a claim about my mental state, even though the Lockean theory assigns a mental state in giving the meaning. So if the treatment of “Yippee” is to assign the mental state of being joyful, then we get a non-cognitivist picture because the mental state assigned to the sentence is not a cognitive mental state. Sentences which are about joy wouldn’t be non-cognitive because the state of joy isn’t assigned to such sentences, rather, beliefs about joy are assigned to such sentences.
I forgot to tie it back in to Berkeley. If Berkeley thinks we can have no idea of joy, then it is hard to see how he could offer a straightforward cognitivist account of the word “joy”. Where does he deny that we can have ideas of passions?
Thanks for the added clarification.
I think you are probably right about the general Lockean picture of meaning.
But the specific debate about whether we have an “idea” of spirit strikes me as a debate about the nature and diversity of mental content as well as the nature and diversity of faculties with which we access this content. I could be wrong about this. But if I’m right, then rejecting the label “idea” in favor of the label “notion” does not get Berkeley very far with respect to issues of meaning. It does, however, press us to think differently about the nature of our acquaintance with certain mental and psychological phenomena. Indeed, Berkeley needs to enrich precisely this feature of ordinary experience if he is to extract concepts such as substance, existence, and agency (or causal power) from “reflexion” on the mind itself.
Peter Browne (in his Procedure of 1728) also takes Locke to task for his indiscriminate use of the term “idea.” Browne argues, perhaps following Berkeley, that we only have “ideas” of sensation and we do not have any “ideas” of mental faculties or operations. His arguments often differentiate (i) the nature of our acquaintance with the mind and its contents from (ii) the nature of our sensory acquaintance with objects in the world. Browne, like Berkeley, appears to be guarding against an impoverished theory of mind to which he believes misuse (or simply overuse) of the term “idea” may lead us.
To answer your question about Berkeley and the passions. Berkeley considers passions in connection with spirit in Principles 27: “the words will, soul, spirit, do not stand…for any idea at all….Though it must be owned at the same time, that we have some notion of soul, spirit, and the operations of the mind, such as willing, loving, hating, in as much as we know or understand the meaning of those words.” However, it is not obvious on the basis of this passage that Berkeley holds a view of the passions like the one I attribute to him.
Also interesting is the fact that Berkeley suggests our understanding of the words is prior to our having the notions (at least evidentially prior).
Where does he suggest this? If that is true, then it certainly supports your reading.
I was perhaps overreading it in the line you quoted:
“the words will, soul, spirit, do not stand…for any idea at all….Though it must be owned at the same time, that we have some notion of soul, spirit, and the operations of the mind, such as willing, loving, hating, in as much as we know or understand the meaning of those words”
Lewis, here is a passage from Richard Burthogge’s Essay (1694) which may be relevant to your project: “the Understanding hath not of its own, (as the Imagination hath) any proper Images, any Figures of the things it converses with, whereby to know or distinguish them; the only Images it has of things (besides those of the Sense, or the Imagination) are the Words which signify them; which do stand therein for the very things themselves. For to give an Example; There is no such thing in the Understanding as an Image, or sensible Figure, of Substance, or of mind, or of matter, or of colour in general, as there is in the sense, or imagination of white, of black, of red, of green, or of other particular Colours: The only Images it has of these, and of all things else that are purely intelligible and mental, are the Words that signify them” (27-8).