Abstract
This paper concerns the phenomenon of pragmatic enrichment, and has a proposal for predicting the occurrence of such enrichments. The idea is that an enrichment of an expressed content c occurs as a means of strengthening the coherence between c and a salient given content c’ of the context, whether c’ is given in discourse, as sentence parts, or through perception. After enrichment, a stronger coherence relation is instantiated than before enrichment. An idea of a strength scale of types of coherence relations is proposed and applied.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
It is a further question whether semantic alone provides us with the interpretation that the two events are contiguous or whether that is a pragmatic choice.
When formulating the first versions of these ideas, in early 2009, I was basically unaware of the coherence theory literature, and only alerted to the connection the following year.
Terminological differences are a little more complex. Officially, explicatures are the results of pragmatic operations, whereas saturations and modulations are the operations themselves, and implicitures are either the operations or the contents added by means of them.
Anaphora resolution, again, is a borderline case; to the extent that coreference is represented syntactically as part of the syntactic input to semantics, pragmatics does not provide a (functional) value.
This is again not the same view as e.g. in Stanley (2000). According to Stanley (2000, pp. 393–394), semantics is concerned with context invariant meaning and the assignment of contextual values to variables in the logical form. Stanley further thinks that semantics in his sense does provide truth conditions of utterances (assertoric content), hence, unlike the present view, that enrichments do not contribute to assertoric content. Moreover, as is demonstrated in Pagin (2005), semantic context sensitivity (in a more neutral sense than Stanley’s) can be encoded by parameters in the meta-language, without relying on variables in the logical form of the object language.
Here a 1 and b 1 correspond to Gephardt and Daschle, respectively, a 2 and b 2, to Gore and the referent of him. q 1 is understood to be the property of being a high-ranking democratic politician. Clearly, the general idea is that there is some clear analogy between objects mentioned in the respective conjuncts, and between the properties, respectively.
This discourse satisfies the relation Inform-accident-and-mention-fruit, proposed by Knott and Dale to illustrate the arbitrariness of unmotivated discourse relations.
Thanks here to Kathrin Glüer.
As pointed out by a referee, we could also have the opposite relation: that John was facing the library explains why he turned left.
For the original proposal of accommodation, concerning presupposition, see Lewis (1979).
Thanks for the example to an anonymous referee.
In the scheme of coherence relations of Sanders et al. (1992, p. 6), the category of causality/implication corresponds to strong connections and the category of addition/conjunction corresponds to weak connections. This is said to be a matter of theoretical intuition.
Leth (2010), chap. 4, argues for applying the theory of discourse relations (rhetorical relations) to pragmatic phenomena, but also for the view that the theory concerns intrinsic properties of texts rather than the psychology of communication. These two ideas are in some tension, it seems to me, and Leth does not try to develop any systematic application to pragmatics.
For propositions we consider a proposition p more complex than a proposition q if the set of p-worlds is a proper subset of the set of q-worlds, i.e. if q follows from p. This is not general enough, however. Between two propositions neither of which entails the other, we would want the one associated with the smaller set to be more complex, but the two sets are likely to have the same infinite cardinality.
See Pagin and Pelletier (2007) for such an account.
Thanks to Jeff King for the suggestion.
The dissociation here between what satisfies coherence (the negated proposition) and what satisfies plausibility (the negative proposition) instantiates a pattern that merits further investigation, but it falls outside the scope of the present paper. Note that negative relations do figure in the Hobbs–Kehler scheme.
As noted above, this is reversed under the scope of negation.
What is the general explanation? According to Carston, the explanation comes from the need of efficiency in processing: “It is just a fact about our minds that we find it easiest to process information when it is sequentially ordered. Imagine that the instructions for assembling a model aeroplane or for knitting a jumper were presented to you in random order or that the frames of a comic strip were not sequentially ordered” (Carston 1995, p. 233).
However, the main contrast is not, as in Carston’s examples, between chronological order and and random order, but e.g. between chronological order and reversed chronological order (then in general read the second conjunct as describing the earlier event). Still, I think it is right that processing efficiency is the key. We read from left to right, and since it is inefficient to backtrack in interpretation, we try to make pragmatic operations on the second conjunct, not on the first after having seen the second.
Also, in general, the presentation of the later event depends on the features of the earlier event. So there are in general pragmatic operations on the presentation of the later event that depend on features of the earlier event. That it is presented later then saves us from backtracking. At least, that would be my attempt at an explanation outline.
Since alternative enrichments are compared and ranked according to availability, it also makes sense to speak of striving for maximal availability.
Hobbs et al. (1993, pp. 80–87), provide an elaborate system for estimating the costs of assumptions in interpretation. They have an informativeness-correctness trade-off for assumptions. The specificity of information should match the linguistic form. For instance, an interpretation that fails to provide a referent for a definite noun phrase is expensive. Longer derivations of logical form are more expensive than short ones. etc.
The total system of Hobbs et al. (1993), within AI, is designed to move from natural language sentences to first-order logical forms with the help of a knowledge base. The main idea is that the truth of the text is explained by the interpretation, typically by means of abduction. In the course of this process, “assumptions are made when necessary” (1993, p. 79). The system is not concerned with enrichments, since enrichments are typically not necessary to render the resulting interpretation true; rather it is typically true before enrichment if true after. Thanks for the reference to an anonymous referee.
Maybe it requires more effort to switch to thinking about B. What would be the reason for believing this?
Several critics (including Bach 2010; Davis 2010) have complained, correctly in my mind, that these principles are not descriptively precise, and hence are weak in predictive power. Wilson and Sperber (2004, pp. 626–627) seem to implicitly acknowledge the point: “For example, the Cognitive Principle of Relevance suggests testable predictions only when combined with descriptions of particular cognitive mechanisms (e.g. for perception, categorization, memory, or inference)”. That is, predictive power is achieved when combined with other theories.
Are the beaches beaches of England or beaches of Australia? According to the experiment reported by Wilson and Matsui, 100 % answer Australia.
Wilson and Matsui (1998, p. 199), criticize Coherence Theory for allowing that two alternative explanations can satisfy the same coherence relation, as in
(35)
a.
I ran to the classroom from the playground. The children were making too much noise.
b.
[\(\ldots\)]. The children [of the playground] were making too much noise [for me to stand].
c.
[\(\ldots\)]. The children [of the classroom] were making too much noise [for me not to interfere].
In this case we can get an explanation of the running whether ‘the children’ refers to the children in the classroom or to the children at the playground. Wilson and Matsui point out that one interpretation or another may be preferred in different circumstances. This is perfectly true, but as long as none of the interpretations is preferred by default, not predicting one reading over the other seems to me a feature, not a bug. Enrichment Theory here predicts ambiguity.
Here I take Levinson to use ‘cohesion’ as equivalent with ‘coherence’. In fact, Sanders et al. (1992, p. 2), and others in the discourse literature, distinguish between the two, taking the cohesion of a text to depend on the linguistic marking of textual relations.
The following is taken (and translated) from the Swedish daily newspaper Dagens Nyheter (DN), 14 July 2013, concerning an event that made the headlines during the work on this paper:
The video that DN published on the web last Friday, showing how a police officer hits a man with a baton, has caused strong reactions. Several members of parliament interviewed by DN think that the police officer uses more violence than required.
The police officer in question was female (obvious from the video), but that is not mentioned anywhere in the article. In some other articles it is revealed only by the use of the feminine possessive pronoun, needed for reporting that the officer’s police dog helped in attacking the man. In yet others it is mentioned explicitly, but only in passing and further down in the article.
Is the paper deliberately playing down the gender aspect, or is it simply treated as irrelevant? A possible explanation for the former would be that the paper wanted to avoid suggesting either that being a woman is a contributing factor in causing the violence (hits a man with a baton [as a female reaction]), or that the transgression is more/less serious if done by a woman (more violence then required [for a female police officer]). If so, the paper used coherence raising considerations.
We can in fact think of predictions that rely on this theory as an empirical test of this part of Coherence Theory.
There are bound to be complications. Here is one. Consider
-
(i)
Adam saw a bird and Bill wanted to buy it.
-
(ii)
Adam saw the bird and Bill wanted to buy it.
The examples are identical except that (i) has an indefinite and (ii) a definite NP. ET predicts that a causal connection will be sought between Adam’s seeing and Bill’s desire. Our background theory tells us that making Bill aware of the bird fits the requirement. In this way Adam causes Bill to know about the bird, and thereby enables him to desire to buy it. The predicted enrichment would be something like: Bill wanted to buy it [after Adam had made him aware of it]. This fits the intuition for (i), but not completely for (ii).
Why? In (ii), the causal interpretation is partly blocked by a Parallel reading: there were two bird-related events; Adam’s seeing it and Bill’s wanting to buy it. ET predicts ambiguity between the Resemblance interpretation and the Possibility interpretation. Then why not ambiguity also in (i)? For some reason the Parallel interpretation appears blocked by the fact that the bird is introduced in the left conjunct by an indefinite NP. This creates an asymmetry that we don’t have in (ii), but why it has this effect on interpretation, I don’t know.
-
(i)
For instance, some theories of attention postulate a cognitive filter: stimuli that pass through the filter are raised to attention (cf. Mole 2009).
Some examples in particular suggest that plausibility operates as a filter even at a conscious level. The following is (summer 2013) written as public notices on subway doors in Athens:
-
(i)
Do not lean against the door. It opens automatically.
(thanks to Kathrin Glüer for the observation). The intuitive enrichment gives the result that the door opens automatically when one leans against it, which is also predicted, since it makes the leaning the cause of the opening. Thereby we have Necessity level coherence. The reading is automatically generated, but consciously discarded: the reader thinks such a mechanism would be criminally insane.
-
(i)
On the question of association vs. inference/computation, Recanati (2004, pp. 28–29) favors association, for completely different reasons. According to Recanati, association is local, and therefore an enrichment made at one point on part of a sentence, does not require computing the semantic value of the entire sentence, the minimal proposition (for a critical discussion of this point, see Pagin 2007, pp. 24–25). Inference, by contrast, would require computing the minimal proposition, according to Recanati.
Sperber and Wilson, by contrast, do regard the comprehension process as inferential, more precisely it is “a dedicated inferential mechanism, a “fast and frugal heuristic,” which automatically computes a hypothesis about the speaker’s meaning on the basis of the linguistic and other evidence provided” (Wilson and Sperber 2004, p. 625).
Indeed they provide the following argument (Wilson and Sperber 2012b): “We will argue that, within the specifically communicative domain, it is indeed rational for hearers to follow a path of least effort in constructing a hypothesis about the speaker’s meaning, and that the pragmatic interpretation process is therefore genuinely inferential [\(\ldots\)].”
I don’t find this very convincing. If for a process to be rational only amounts to being successful, there is no reason why it should involve inference. And if rationality requires doing something for a reason, then this seems to incur an extra cost. If I do ϕ for the reason that doing ϕ would be to follow the path of least effort, then it seems I am wasting energy on an inference that gives the same result as just following the path of least effort without a reason. Hence, because of the extra reasoning effort, I am not following the path of least effort. What rationality amounts to here is unclear.
References
Asher, N., & Lascarides, A. (2003). Logics of conversation. Studies in natural language processing. Cambridge: Cambridge University Press.
Bach, K. (1994). Conversational impliciture. Mind and Language 9, 124–162.
Bach, K. (2010). ‘Impliciture vs. explicature: What’s the difference?’, In B. Soria, E. Romero (Eds.), Explicit communication. Robyn Carston’s pragmatics, (pp. 126–137). London: PalgraveMacMillan.
Carston, R. (1995). Quantity maxims and generalized implicature. Lingua 96, 213–244.
Carston, R. (2002). Thoughts and utterances. Oxford: Blackwell.
Carston, R. (2004a). Relevance theory and the saying/implicating distinction. In L. R. Horn, & W. Gregory (Eds.), The handbook of pragmatics (pp. 633–656). Oxford: Blackwell.
Carston, R. (2004b). ‘Truth-conditional content and conversational implicature’. In B. Claudia. (Eds.), The semantics/ pragmatics distinction (pp. 65–100). Stanford, CA: CSLI.
Davidson, D. (1984). Inquiries into truth and interpretation. Oxford: Clarendon Press.
Davis, W. (2010). Implicature. http://plato.stanford.edu/archives/win2010/entries/implicature/.
Dowty, D. (1986). The effects of aspectual class on the temporal structure of discourse: Semantics or pragmatics? Linguistics and Philosophy 9, 37–61.
Grice, H. P. (1975). ‘Logic and conversation’. In P. Cole, J. L. Morgan (Eds.), Speech acts, Vol. 3. Syntax and semantics. New York: Academic Press. Reprinted in Grice 1989, chap. 2, pp. 22–40. Page references to the reprint.
Grice, H. P. (1989). Studies in the ways of words. Cambridge, MA: Harvard University Press.
Groenendijk, J., & Stokhof, M. (1994). Questions. In A. ter Meulen, & J. van Benthem (Eds.), Handbook of logic and language, Chap. 19. (pp. 1055–1124). Amsterdam: Elsevier.
Hobbs, J. R. (1985). On the coherence and structure of discourse. Technical report CSLI-85–37. Stanford University, CSLI.
Hobbs, J. R., et al. (1993). Interpretation as abduction. Artificial Intelligence 63, 69–142.
Horn, L. R. (2004). ‘Implicature’. In L. R. Horn, G. Ward (Eds.), The handbook of pragmatics, (pp. 3–28). Oxford: Blackwell Publishing.
Horn, L. R., & Ward, G. (Eds.) (2004). The handbook of pragmatics. Oxford: Blackwell.
Hume, D. (1748). An inquiry concerning human understanding, 1955 Edn. New York: Liberal Arts Press.
Kamp, H., & Reyle, U. (1993). From dscourse to logic: Introduction to model theoretic semantics of natural language, formal logic and discourse representation theory. Dordrecht: Kluwer.
Kaplan, D. (1989). Demonstratives. In J. Almog, J. Perry, H., & Wettstein (Eds.), Themes from Kaplan, (pp. 481–563). Oxford: Oxford University Press.
Kehler, A. (2002). Coherence, reference, and the theory of grammar. Stanford University: CSLI.
Knott, A., & Dale, R. (1994). Using linguistic phenomena to motivate a set of coherence relations. Discourse Processes 18, 35–62.
Leth, P. (2010). `Paraphrase and Rhetorical Adjustment. An Essay on Contextualism and Cohesion. Ph.D. thesis. University of Gothenburg and Université Paris-Est Créteil, Gothenburg.
Levinson, S. C (2000). Presumptive meanings. The theory of generalized conversational implicature. Cambridge, MA: MIT Press.
Lewis, D. (1979). Scorekeeping in a language game. Journal of Philosophical Logic 8. Reprinted in Lewis 1983, 233–49. Page references to the reprint, pp. 339–359.
Lewis, D. (1983). Philosophical papers, Vol. 1. Oxford: Oxford University Press.
Maienborn, C., von Heusinger, K., & Portner, P. (Eds.). (2011). Semantics. An international handbook of natural language meaning. Berlin: Mouton de Gruyter.
Mole, C. (2009). Attention. http://plato.stanford.edu/archives/fall2009/entries/attention/.
Nunberg, G. (1979). The non-uniqueness of semantic solutions: Polysemy. Linguistics and Philosophy 3, 143–184.
Nunberg, G. (1993). Indexicality and deixis. Linguistics and Philosophy 16, 1–43.
Nunberg, G. (1995). Transfers of meaning. Journal of Semantics 12, 109–132.
Ogihara, T. (2011). Tense, Chap. 57. In C. Maienborn, K. von Heusinger, P. Portner (Eds.), Semantics. An international handbook of natural language meaning (Vol. 2, pp. 1483–1464). Berlin: Mouton de Gruyter.
Pagin, P. (2005). Compositionality and context. In Gerhard, P., Georg, P. (Eds.), Contextualism in philosophy (pp. 303–348). Oxford: Oxford University Press.
Pagin, P. (2007). Pragmatic composition? In M. Zouhar, T. Marvan (Eds.), Svet jazyka a svet za jazykom (pp. 11–26). Supplement volume of Organon F.
Pagin, P., & Pelletier, F. J. (2007). Content, context and communication. In G. Preyer, G. Peter (Eds.), Context-sensitivity and semantic minimalism. New essays on seman tics and pragmatics (pp. 25–62). Oxford: Oxford University Press.
Portner, P. (2011). Perfect and progressive, Chap. 49. In C. Maienborn, K. von Heusinger, P. Portner (Eds.), Semantics. An international handbook of natural language meaning (Vol. 2, pp. 1217–1261). Berlin: Mouton de Gruyter.
Quine Van, W. O. (1960). Word and object. Cambridge, MA: MIT Press.
Recanati, F. (2004). Literal meaning. Cambridge: Cambridge University Press.
Recanati, François (2010). Truth-conditional pragmatics. Oxford: Oxford University Press.
Reichenbach, H. (1947). Elements of symbolic logic. New York: Macmillan
Rumelhart, D. (1979). Some problems with the notion of literal meanings. In A Ortony (Ed.), Metaphor and thought. 2nd ed, (pp. 71–82). Cambridge: Cambridge University Press.
Sanders, T. J. M., Spooren, W. P. M., & Noordman, L. G. M. (1992). Toward a taxonomy of coherence relations. Discourse Processes 15, 1–35.
Sperber, D., & Wilson, D. (1986). Relevance. Communication and cognition, 1st edn. Oxford: Blackwell.
Sperber, D., & Wilson, D. (1995). Relevance. Communication and cognition, 2nd edn. Oxford: Blackwell.
Stanley, J. (2000). Context and logical form. Linguistics and Philosophy 23, 391–434.
Wilson, D., & Matsui, T. (1998). Recent approaches to bridging: Truth, coherence, relevance. In UCL Working Papers in Linguistics 10, pp. 1–28. Reprinted in Wilson and Sperber 2012a, pp. 187–209. Page references to the reprint.
Wilson, D., & Sperber, D. (2004). Relevance theory. In Laurence R. Horn, G. W. (Eds.), The handbook of pragmatics (pp. 607–632). Oxford: Blackwell Publishing.
Wilson, D., & Sperber, D. (2012a). Meaning and relevance. Cambridge: California University Press.
Wilson, D., & Sperber, D. (2012b). ‘Pragmatics, modularity, and mindreading’, Chap. 12. In meaning and relevance (pp. 261–278). Cambridge: California University Press.
Zobel, S. (2010). Non-standard uses of German 1st person singular pronouns. In K. Nakakoji, Y. Murakami, & E. McCready (Eds.), JSAI-isAI, LNAI 6284 (pp. 292–311). Berlin: Springer.
Acknowledgments
Earlier versions of this work have been presented at a conference on Peter Strawson’s work at the Academy of Science in Prague in April 2009, a conference on systematic pragmatics at Stockholm University in May 2009, at the Euro-XPrag conference in Leuven in June 2010, at the Content, Context and Conversation Lichtenberg-Kolleg Workshop at the Georg-August Universität Göttingen in June 2011, and at the Language and Linguistics workshop at the Inter-University Center of Dubrovnik in September 2011. I am grateful for comments from the audiences on those occasions, especially Nicholas Asher, Christian Beyer, Stephen Butterfill, Herman Cappelen, Robyn Carston, Mari Duží, Bart Geurts, Kathrin Glüer, Julie Hunter, Jeff King, Luisa Martí, Petr Kotatko, Stephen Neale, Ira Noveck, Stefano Predelli, Francois Recanati, Petra Schumacher, Mandy Simons, Barry Smith, Paul Snowdon, Dag Westerståhl, and Deirdre Wilson. As always, I owe much to Kathrin Glüer’s inputs in discussions over the years of working on this theme. I am also extremely indebted to comments from two anonymous referees. Their remarks led to improvements not only of the presentation and the discussion, but also of the theory itself. This work has been made possible by a Grant from the Tercentenary Foundation of the Swedish National Bank, project Interpretational Complexity, as well as by support from the ESF Eurocores project CCCOM, Communication in Context, led by Professor Åsa Wikforss.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Pagin, P. Pragmatic enrichment as coherence raising. Philos Stud 168, 59–100 (2014). https://doi.org/10.1007/s11098-013-0221-8
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-013-0221-8