Keith DeRose, on Single Scoreboard Semantics

Keith DeRose (2004) describes a conversation in which two interlocutors disagree about whether one of them knows something. The first argues that the second doesn't know something, but the second insists that she does. Their ability to convince each other of the truth of their claims may be determined by how many "points" they score in a "language game." For agreement to be achieved however, they must to some extent share the same "conversational scoreboard," and it may be more difficult for them to agree if they each have their own personal scoreboard. If they don't share the same understanding of what it means to "know" something, then they may not be able to agree on whose claims to knowledge are true.

      Single scoreboard semantics, according to DeRose, proposes a single scoreboard for a given conversation, on which the knowledge claims of each interlocutor can be recorded. He explains that epistemic contextualism requires that the knowledge claims of each interlocutor be evaluated according to their context. If the first interlocutor has established a higher standard for what it means to "know" something than has the second interlocutor, then the first may be correct in saying that the second doesn't actually know what she says she knows. Epistemic invariantism, on the other hand, is the position that the truth or falsity of knowledge claims doesn't depend on the epistemic standard of the context (whatever is true in a low standard context will also be true in a high standard context).   

      DeRose supports the contextualist position that the truth or falsity of knowledge claims varies according to the epistemic standard of the context. He notes that his use of the term "conversational scoreboard" differs from that of David Lewis (1979), insofar as Lewis uses the term to describe the mental scoreboards of each interlocutor (who may not agree on the actual score), while DeRose uses the term to describe a single scoreboard that by definition gives the right score.¹

      According to Lewis, there may be constitutive as well as regulative rules regarding how a language game is played and how the score is recorded. If each interlocutor has their own mental scoreboard, then for each of them the score may be whatever their mental scoreboard says it is. But score keeping is also governed by rules, so the interlocutors may also sometimes disagree if one of them has broken the rules. On the other hand, they may also accommodate each other by allowing for some flexibility and variation in the interpretation of the rules, so that agreement can be achieved.

      DeRose therefore doesn't address the question of whether it is or isn't the case that there are sometimes or always multiple personal (or mental) scoreboards in a given conversation. Presumably, the mental scoreboards of each interlocutor must agree to some extent or at least partially coincide if the interlocutors are to agree on the comparative validity of each other's knowledge claims. But to what degree must they agree? 

      Conversational markers of agreement, such as "Yeah," "Sure," "Uh huh," "I see," "okay," and "I suppose so," may enable interlocutors to recognize that they are in fact sharing the same or similar conversational scoreboards. Interlocutors may also employ a variety of conversational strategies (such as back channel utterances, response tokens, requests for clarification, continuers, repetitions, collaborative finishers, and nonverbal responses) in order to manage and support communication.

      DeRose addresses the case in which the conversation between two interlocutors is observed or monitored by a third party who keeps their own scoreboard. In this case, it may not matter whether each interlocutor has their own personal scoreboard, as long as they agree to arbitration by the (presumably neutral, fair, and impartial) third party. Disagreement can then be resolved by a "binding arbitration" model of conversational score keeping, in which the third party decides which interlocutor's arguments are more compelling and reasonable.

      DeRose also describes the case of the "exploding scoreboard," in which there's no correct conversational score and the knowledge claims of each interlocutor cannot be judged as true or false.

      He also considers the case in which interlocutors disagree and there's a divergence or  "gap" in the truth conditions of their relevant claims. This "gap view" may help to resolve disagreement between interlocutors when there's vagueness or uncertainty in their knowledge claims.

     Would the single scoreboard versus multiple scoreboards model be subject to question if one of the interlocutors simply imposed their knowledge claims on the other by force, threat, or intimidation? Perhaps in this case there wouldn't be any social space for disagreement.

      Verena Gottschling (2004) asks what happens when there's a change in epistemic standards during the course of a conversation and only one of the participants accepts the change. If only one participant agrees with the method of score keeping, then how do we evaluate the knowledge claims of the other participants? Gottschling explains that Lewis, in his article "Elusive Knowledge" (1996), says that the epistemic standards of a conversation may be lowered if one of the participants says something that's true under a lower epistemic standard and the other participants don't disagree. On the other hand, the epistemic standards of a conversation may be raised if one of the participants calls the attention of the others to some remote possibility that they must account for because of the context, such as when jurors must decide whether a defendant is guilty beyond a reasonable doubt. 

      Gottschling criticizes the notion that conversational participants can each have their own personal scoreboard, because then our intuition that their views may actually contradict each other seems to be violated. She also criticizes the notion that there can be a single scoreboard, because then our intuition that each participant may actually have their own persisting individual standards seems to be violated. She therefore concludes that we should reexamine our intuition of contradiction, since contextualism should be understood as recognizing that a change in the content of knowledge claims by conversational participants may cause us to feel their claims contradict each other when in fact they do not.²

FOOTNOTES

¹Keith DeRose, "Single Scoreboard Semantics," in Philosophical Studies, Volume 119 (1-2), 2004, p. 19.

²Verena Gottschling, "Keeping the Conversational Score: Constraints for an Optimal Contextualist Answer?", in Contextualisms in Epistemology, edited by Elke Brendel and Christoph Jäger (Dordrecht: Springer, 2005), p. 168.

OTHER SOURCES

David Lewis, "Scorekeeping in a Language Game," in Journal of Philosophical Logic, 8, 1979, pp. 339-359.

David Lewis, "Elusive Knowledge," in Australasian Journal of Philosophy, 74, 1996, pp. 549-567.