Steffen Koch (University of Bielefeld), "Merely verbal agreement, speaker-meaning, and defective context"

Synthese, 2025

Merely verbal agreement, in philosophy and beyond

By Steffen Koch

Everyone has had the experience of talking past each other. You're using the same words as the person you’re talking to, but nonetheless, the two of you mean quite different things. If you're struggling to find examples, just think of discussions about grand issues like freedom of speech, democracy, or free will. When discussing such issues, it is very easy to talk past each other. You both think that you use “democracy” in the same sense, and that the issue is whether India, for example, counts as a democracy; but in fact part of the issue is not about India at all, but about what each of you means by “democracy.”

Talking past each other can have several consequences. It can be the source of what philosophers like David Chalmers call a “merely verbal dispute”: a situation in which the participants of a conversation think they are having a meaningful, subject-related dispute about some issue, whereas in reality, their dispute is largely due to different understanding of some of the key concepts or terms. But talking past each other can also be the source of merely verbal agreement: the parties appear to agree on some issues, but their agreement is based on different understandings or conceptions of the expressions they used.

Philosophers devote a great deal of attention to merely verbal disputes. They discuss what exactly it means to talk past each other in the relevant sense. They also discuss what kinds of philosophical disputes are likely candidates for merely verbal disputes, with candidates ranging from debates about material composition, endurantism, reductionism, free will, knowledge and justification, or the semantic vs. pragmatic distinction. By contrast, philosophers have largely ignored the related phenomenon of merely verbal agreement. Given the apparent pervasiveness of disputes and disagreements among philosophers, this one-sided interest in disputes can perhaps be excused – but it is still a mistake, for merely verbal agreement is likely to be just as pervasive as merely verbal disputes, both in philosophy and in everyday life, and its effects may be even more severe.

The first step to better understanding and evaluating verbal agreements is to get clearer about what they actually amount to. Before we do that, however, a word about terminology: When I speak of “agreement” or “disagreement,” I am referring to a state of congruent or incongruent doxastic attitudes. The behavioral counterpart of disagreement is a dispute: a dialogical episode in which parties express conflicting views about an issue. Unfortunately, the English language does not provide us with a behavioral counterpart of agreement. To fill this linguistic gap, I propose to use the term “accord” as a technical term to refer to a dialogical episode in which parties verbally express converging attitudes.

My first key claim is that we should conceive of merely verbal agreement and accord as pragmatic phenomena. What I mean by this is best brought out by an example. Suppose that two philosophers, McA and McB, talk about free will. McA says: “I think people have free will,” to which McB responds “I agree.” Suppose, however, that what McA means by their utterance is that people have the ability to deliberate about their actions and to execute the result of this deliberation (for short, compatibilist free will), whereas what McB means to agree with is that people have the ability to act in ways that are not determined by the laws of nature (for short, incompatibilist free will). McA does not believe that people possess incompatibilist free will. McA and McB are having a merely verbal agreement.

David Chalmers has suggested that merely verbal disputes are due to underlying disagreements about the meaning of some of the terms involved in the dispute. Transposed to merely verbal agreement, this would suggest that McA and McB are having a merely verbal agreement to the extent that they disagree about what “free will” means. But this cannot be true. For all we know, McA and McB might agree that “free will” standardly means compatibilist free will. As long as what McB intends to agree with, in this particular situation, is whether people possess incompatibilist free will, they end up having a merely verbal agreement regardless. People can and often do use terms in ways that differ from what they believe that the term standardly means. These cases demonstrate that what really drives merely verbal agreement is what the parties mean to say (what philosophers call “speaker’s meaning”), rather than what they believe certain terms standardly mean.

In light of this, I propose the following definition of a merely verbal agreement:

Two parties A and B have a merely verbal agreement over A’s assertion S iff

(i) by uttering S, A asserts that p;

(ii) B thinks that, by uttering S, A asserts that q (q ¹ p);

(iii) B believes q but does not believe p.

The basic idea here is simple enough: A and B are having a merely verbal agreement just in case there is a discrepancy between what A asserts and what B thinks that A asserts, such that B agrees with what they think that A has asserted, but not with what they have actually asserted. Crucially, in line with the above, the content of an assertion is understood in terms of speaker’s meaning rather than as what the sentence used to make the assertion literally means in the common language. Moreover, following Robert Stalnaker, we can understand assertions as proposals to add their content to the conversational common ground: to the set of propositions that all participants of a conversation regard as mutually accepted. Such proposals can be accepted or rejected by an interlocuter. If they are accepted, they become part of their common ground, if not, they don’t. With these notions in the background, we can define merely verbal accord as follows:

Two parties A and B are having a merely verbal accord over A’s assertion S iff.

(i) by uttering S, A asserts p;

(ii) B thinks that, by uttering S, A asserts q (q ¹ p);

(iii) by accepting S, B means to accept q;

(iv) A thinks that, by accepting S, B means to accept p.

Rather than go into more detail about the definitions, let us dwell a bit on the potential consequences of a merely verbal agreement. I think there are at least three noteworthy consequences. The first is the most obvious: merely verbal agreement leads to mistaken beliefs about the common ground of the conversation. Each party thinks that the proposition they have uttered has been accepted by the other, but neither has, and so neither is actually part of their common ground. To borrow a term from Andrew Peet, this means that their common ground is opaquely defective. Notably, the much more discussed counterparts of merely verbal agreement – merely verbal disputes – do not have this consequence. While merely verbal disputes keep the common ground from growing, they do not introduce defects (at least not of the sort we are concerned with here – details in the paper).

The second reason to care about merely verbal agreements is that they sometimes impede cooperative action. Consider the following example: Fred and Susan are researchers, and they plan to co-host a lecture series. Fred is responsible for sending out invitations to speakers and other attendees. While discussing the frequency of their meetings, Susan says, “Let's meet biweekly,” to which Fred replies: “All right.” Unbeknownst to either of them, they are having a merely verbal accord: What Susan means to assert is that they should meet every other week, but what Fred understood and agreed to is that they should meet twice a week. Suppose further that their dialogue takes place in a context where both options seem initially plausible, and that Fred has no other information about Susan's meeting preferences.

What is going to happen next? Well, reality is messy, so nobody knows exactly, but here is something that seems not entirely unlikely: Fred goes home thinking that they have agreed to meet twice a week. Consequently, he schedules two sessions a week. Neither of them notices the misunderstanding until its real-world consequences begin to unfold – for example, when Susan receives invitation mails for two meetings per week, or, even worse, when Fred asks her why she did not appear at one of their meetings. At this point, however, damage is already done. Note also that merely verbal disputes do not have this effect. If Fred and Susan had had a merely verbal dispute about the meeting frequency, Fred would not have send out invitations before discussing the issue further.

The third and final reason to be concerned about merely verbal agreements is that they are detrimental to philosophical progress. This is why it is so tempting to think that philosophy is full of merely verbal disputes: Philosophers often argue about whether a proposition is true, while holding different views (or theories, or conceptions) about key concepts involved in that proposition. For example, they dispute whether humans have free will, while disagreeing about what it means to have free will; they dispute whether philosophers use intuitions as evidence, while disagreeing about what intuitions are; they dispute whether there is an analytic/synthetic distinction, while disagreeing about what that distinction even means; they...you get the point.

Interestingly, however, this same reasoning extends to merely verbal agreement: If two people are in accord that some x is F, while at the same time disagreeing about what it even means to be F, then chances are that their accord is merely verbal. Now, those who have neglected merely verbal agreement and accord might want to defend themselves by pointing out that in philosophy (apparent) agreement is a rather rare phenomenon, so that the possibility of merely verbal agreement can safely be ignored. But this is clearly a lame argument. In fact, philosophers seem to agree on many things - the fact that disputes and disagreements are naturally louder and more visible should not blind us to this fact. Note that even if philosophers are split 50/50 on a question, such as whether we have free will, there is still a lot of apparent agreement among those in the same camp.

If the foregoing is true, we should expect philosophy to contain at least as much merely verbal agreement as it does merely verbal disagreement. Why does this matter? Professional philosophy is often organized as a kind of collaborative action: we organize events together, write papers together, make professional decisions together, and so on. This means that merely verbal agreement will have the same effects in these areas of philosophy as it does in other areas of our lives. But there is also an arguably worse, or at least more interesting, effect: Philosophical progress happens when we address and ultimately resolve our disagreements in a constructive and truth-oriented spirit, by exchanging arguments for or against our respective views, by considering the implications of our opponents' views, etc. Merely verbal agreement functions as a cover for disagreement: it makes us think that we agree on issues about which we actually disagree, thereby depriving us of the opportunity to resolve those disagreements constructively. Merely verbal agreement is thus a hindrance to our collective search for truth. If that's not reason enough for a philosopher to be interested in merely verbal agreement, what is?

The paper: https://philpapers.org/rec/KOCMVA