Normative Semantics: What Should Our Words Mean?…

Parts of this post draw from David Chalmers’ paper “Verbal Disputes“* (a lecture on the same topic can be found here.) , I will move from this discussion to my original point. Quotations in this post all refer to this paper.

Chalmers draws a distinction between ‘verbal’ and ‘substantive’ philosophical disputes. After consideration of multiple suggestions, Chalmers settles on this definition for a “broadly verbal dispute”:

A dispute over S is (broadly) verbal when for some expression T in S, the parties disagree about the meaning of T, and the dispute over S arises wholly in virtue of this disagreement regarding T.

The way I would phrase the definition of a merely verbal dispute is: Conversational emphasis on semantic concordance appears sufficient for dispute resolution. In other words, if we satisfy each other that we are using the relevant terms in a similar enough way, does any disagreement about some claim(s) remain? If not, we have discovered a region of merely verbal dispute. Below, I will be offering some suggestions for a community might do when they arrive at this situation. But first, let’s hear Chalmers help us understand the playing field.

To elucidate, Chalmers uses a quaint example from William James, wherein “A man walks rapidly around a tree, while a squirrel moves on the tree trunk. Both face the tree at all times, but the tree trunk stays between them”. In a case such as this, a dispute may arise between interlocutors as to whether the man goes “round” the squirrel, or not. James uses this as a paradigm case of a ‘merely verbal’ dispute and deflates thusly:

Which party is right depends on what you practically mean by ‘going round’ the squirrel. If you mean passing from the north of him to the east, then to the south, then to the west, and then to the north of him again, obviously the man does go round him, for he occupies these successive positions. But if on the contrary you mean being first in front of him, then on the right of him then behind him, then on his left, and finally in front again, it is quite as obvious that the man fails to go round him… Make the distinction, and there is no occasion for any farther dispute. (James, Pragmatism, 1907, as quoted by Chalmers)

To plug this situation into Chalmers’ definitional formula, S = The Claim: “The man goes round the squirrel” and T = The Expression: “goes round”. Because, counterfactually, both disputants would agree to the Claims S1 and S2 were James to intervene and provide this linguistic clarification for the Expression T– this counts as a ‘merely verbal’ dispute. Both interlocutors would say something like, ‘Yes, alright, if “going round” means the former then S1… and if “going round” means the latter then S2… We agree.’

One sensible methodology for establishing the verbal nature of a dispute, suggests Chalmers, is “the method of elimination”. If, for some reason or other, a disputant (or external adjudicator) suspects the presence of a ‘merely verbal’ dispute, they may eliminate the suspicious linguistic object from the sentence in question and examine whether or not a dispute remains.

Chalmers moves to consider a more philosophically robust example:

The method of elimination can be applied to many disputes in philosophy. To illustrate a possible use, I will start with an issue that has often been accused of giving rise to verbal disputes, and in which proponents are relatively sophisticated about these issues: the question of free will and determinism.

Suppose that a compatibilist says ‘Free will is compatible with determinism’, and an incompatibilist says ‘No, free will is not compatible with determinism’. A challenger may suggest that the dispute is verbal, and that the dispute arises only because the parties mean different things by ‘free will’.

We can then apply the method of elimination: bar the term ‘free will’, and see whether there are residual disputes.

After elimination, the disputants would be asked to consider the Claim formula ‘Is X compatible with determinism’ [CX] and each could provide their own account of X/’free will’ and see if a dispute remains. To use Chalmers’ examples, say that Susan defines ‘free will’ as “the ability to do what one wants” [D1] while Charlie defines it as “the ability to ultimately originate one’s choices” [D2]. It may turn out that the disputants learn something by alternately plugging in [D1] and [D2] for X and re-examining the agreement status of the Claims [C1] and [C2]. Perhaps both agree that [D1] is compatible with determinism but that [D2] still is not. We would then have discovered that at least part of the dispute here was verbal, and refined the grounds of the debate.

I agree with Chalmers that some [I would say ‘many’] philosophical disputes either are merely verbal, or at least have a verbal component in the sense described. That apparent disputes about the world can dissolve merely through a focusing on semantics, as with James’ “round the squirrel”. I agree that people ought to be vigilant for situations of this type and strive to minimize episodes of mistaking a ‘verbal’ dispute for a ‘substantive’ one.

But, I would like to ask people to consider this possible response to the question: Then what? Say some community resolves [or dissolves] a putative dispute through the method of elimination– this is a positive step forward and I’m all for it –yet I find it insufficiently successful at the ultimate goal of reaching genuine dispute resolution and agreement. I claim that ‘reasonable people cannot disagree’ and a semantic disagreement is still disagreement. Yes, let us not mistake verbal disputes for substantive ones… But I would like to add: Let us not rest there, let us transform verbal disputes into substantive disputes and resolve those. How? By the introduction of a new scientifico-philosophical practice we might call normative semantics.

What if it were a domain of inquiry to examine questions like ‘What should words mean?’, or ‘How will we, as some linguistic community, choose to define and agree to use this word going forward?’

As usual, let’s begin attempting to legitimize this suggestion by pointing out some of what I take to be part of a pretty wide agreement base:

It seems to me that if one is willing to buy those premises, that it doesn’t take much more to establish the legitimacy of normative semantics. Just add:

And the conclusion, as far as I can see, follows:

If one is with me so far, I think at this point we have established an argument that the project of normative semantics is both possible and motivated. But we have not yet quite reached a destination where my desired claim ‘Word w should have definition d‘. Let’s see if we can make any headway there.

I take it as relatively uncontroversial that 1) there are ‘words’, and 2) lexicography is a more or less ‘respectable’ science– that one can discover sufficient consensus about the linguistic practices of a community to develop a suggestive, interconnected record of word usage as an approximate pre-theoretical natural language semantics. Contemporary high-quality lexicographical record is the first resource of normative semantics. Stated simply: What does our best science think these words mean now?

Next consider ‘conceptual schemes’ or ‘world-views’– abstract webs of theoretical posits, engineered through some process or other, that allows an agent or group to categorically interpret a data set (e.g. ‘sense experience’). One might call them, abstraction patterns. It appears that in order to function as the sorts of animals we are, that we are required to abstract constantly. We must ‘forget’ or neglect wide swaths of the available incoming data, we must cognitively categorize to flexibly but reliably behave in our real time environmental struggles.

Then, mix in the idea that these agents or groups of agents have projects and concerns; and that these concerns, to some degree, influence the design of the conceptual scheme and vice verse. Again, it appears uncontroversial to me, that given any project set and population, an indefinite number of cognitively actualizable conceptual schemes could be instantiated which themselves could be hierarchically ordered by an empirical examination of project-progress efficacy.

Give a community with project-set {P} conceptual schemes C1…Cn and measure the energy expenditure as they invest in completion of {P}. Though there may be ties, and there may be no scheme which is ‘perfect’, it seems likely that some of the schemes will function more efficiently than others.

Now, combine these points with a boring empirical fact about us earth bound human beings in 2018 that one of the most common ways we interact with nodes in conceptual schemes is through the lever of natural language ‘word’ use. We ‘talk’. We express conceptual abstract ideas, among other ways, through the activity of ‘small mouth noises’, or intricate calligraphy. There are obvious and uncontroversial, if commonly unnoticed, norms on words already due to what we might call ‘agentive situational facts’– e.g. no word in English is longer than it would take a normal speaker to pronounce in an hour.

With these additional considerations in place, I feel able to (re)state the point I’d like you to consider. Since we use conceptual schemes to interpret the world for the purpose of coordinating our our project-directed energy expenditures, and we use words to ‘interact’ with the concepts both personally and socially, it seems to me that some word-to-concept mappings would be more efficacious at achieving some projects.

Normative semantics would be the process of examining at least 1) contemporary and historical lexicography, 2) apparently relevant segments of conceptual schemes, and 3) communal agentive situational facts to make suggestions about what e.g. professional philosophers ought to mean when they utter a sentence containing, say, a token of ‘free will’.

Part of the answer to that question, ‘What should we take “free will” mean?’ will involve facts about current usage in that speaker’s community, facts about what concepts that community wants words for, facts about what the agents involved in that conversation are able and willing to do with words, and likely many other factors. Examinations of this type would constitute the project I’m calling ‘normative semantics’, an attempt to ‘substantivize’ ‘merely verbal’ disputes. ‘Well, given what we want to accomplish here, where we’re starting from, who we are, etc. how do we as a community want to agree to employ this word / phrase going forward?’ To me, this seems like a useful activity and plays a relatively major role in my metaphilosophical and methodological musings.

If more people took normative semantics seriously, we could attempt substantive, scientific, arguments about what words should mean, and not rest comfortably with James/Korzybski/Chalmers indexing deflations. One advantage the normative semantics program would have over the deflationary, I think, is that through this method we could diminish the number of ‘verbal’ disputes in the first place! If the community agreed on what words and phrases should mean, it would be less likely that a disputant would arrive with semantic idiosyncrasies and our communal argumentative projects would likely proceed more efficiently.
————
* A Brief Aside: It strikes me as a commentary on modern academia’s strange ignorance of Korzybskian General Semantics, that a paper which includes lines such as “To adjudicate whether this dispute is verbal, one can bar ‘physicalism’, introduce ‘physicalism1’ and ‘physicalism2’, and see whether there is a residual disagreement.” as though it were a novel suggestion, does not cite Science and Sanity. This appears to me a precise rediscovery of the General Semantic practice of ‘indexing’ that Korzybski wrote of almost 100 years ago.