Michelle Liu is currently a Leverhulme Early Career Fellowship at the University of Hertfordshire and will take up a permanent position at Monash from July 2023. She works on various topics in philosophy of mind, philosophy of language and aesthetics.
A post by Michelle Liu
This blogpost, as its title suggests, is about the role of mental imagery in comprehending language. Here, mental imagery is understood as perceptual simulation or perceptual representation without direct external stimulus.
It is widely known that mental imagery plays a crucial role in understanding novel and poetic language. The comprehension of some metaphors, as philosopher Mitchell Green (2017) argues, requires the construction of conscious mental imagery. Green calls them ‘image-demanding metaphors’. As an example, he tells the story of Wittgenstein’s first meeting with Frege, about which Wittgenstein recalled that Frege had ‘wiped the floor with’ him. As someone who had never heard of this phrase, Green (2017: 34) notes that only after having formed a mental image of ‘one person using another to sweep or mop a floor’, had he understood Wittgenstein’s point that he was intellectually dominated by Frege.
Relatedly, writers and poets often rely on mental imagery to communicate certain feelings or sentiments that are otherwise hard to put into words (see Liu forthcoming). In I Felt a Funeral, in My Brain, Emily Dickinson articulates an intense feeling of loss and suppression through various sensory aspects of a funeral, e.g. the footsteps of mourners, the beating of a drum, the sound of lifting the coffin, the procession of heavy boots, and the ringing of bells. In experiencing mental imagery associated with such sensory descriptions, the reader gains a purchase on the overwhelming feeling intended by Dickinson and may come to understand, to quote linguist Robyn Carston (2010:311), ‘how the world looks and feels to the poet in (her) exalted state of mind’.
Mental imagery is not only crucial to understanding novel and poetic language, recent psycholinguistic literature suggests that it is also important for comprehending ordinary, literal language. In his fascinating book Louder Than Words, psychologist Benjamin Bergen (2012) makes a convincing case that in many situations, ‘we understand language by simulating in our minds what it would be like to experience the things that the language describes’ (Bergen 2012: 13). Various experiments conducted by psychologist Rolf Zwaan and his team have provided us with abundant evidence that language users activate visual imagery of various features of the objects described in language, including object’s orientations (Stanfield and Zwaan 2001), shapes (Zwaan et al. 2002), spatial relations (Zwaan and Yaxley 2003), motions (Zwaan et al. 2004), and colours (Richter and Zwaan 2009).
The precise content of mental imagery activated by a sentence depends on a number of things. Whereas the semantic contents of words determine what is simulated or imagined, grammatical features of a sentence can modulate what perspective to take, what part of some mental imagery to focus on, and the level of detail with respect to its components (Bergen 2012: 118). For instance, language users tend to take a third-person observer perspective when comprehending sentences in the third-person (e.g. ‘He is slicing a tomato’); they tend to take a first-person participant perspective, identifying themselves with the person who is speaking, if the sentence uses the first person ‘I’ and contextual information about the ‘I’ is kept to a minimum (e.g. ‘I am slicing a tomato’) (Brunyé et al. 2009; Bergen 2012: 110-114). Research has also shown that sentences in the progressive tense (e.g. ‘John is closing the drawer’) allow greater activation of mental imagery of the described event compared to sentences in the perfect tense (e.g. ‘John has closed the drawer’), which tend to prompt language-users to activate mental imagery of the end state of an event (Bergen and Wheeler 2010).
This simulation view of language processing, where the latter is seen as a dynamic process that frequently involves the construction of mental imagery, departs from the traditional view where language processing is thought to be a matter of abstract symbol manipulation independent of the brain’s perceptual-motor systems (Hauser, Chomsky and Fitch 2002). In addition to being empirically supported, however, it holds explanatory potential for linguists and philosophers. In my own work (see Liu 2022), I have focused on how mental imagery can shed light on the processing of polysemous words, i.e. words with multiple distinct but related senses.
Here, I will focus on the phenomenon of copredication, which is thought to be associated with a specific subset of polysemous words. Consider:
(1) The book is thick and interesting.
(2) The lunch was delicious but took a long time.
(3) The school caught fire when celebrating students’ graduation.
(4) The window was broken many times and had to be boarded up.
In all these cases, the noun in the sentence seems to be used in different senses. In (1), it is book-as-tome which is thick, and book-as-content that is interesting. In (2), lunch-as-food was delicious and lunch-as-event took a long time. In (3), it is school-the-building that caught fire and school-as-staff-and-students that celebrate the graduation. In (4), window-as-glass was broken and window-as-opening had to be boarded up. In the literature, it is common to treat nouns in (1)-(4) as polysemous – they are examples of what is called ‘inherent polysemy’. Such polysemes permit coordination of different senses of the noun (e.g. Pustejovsky 1995; Asher 2011; Falkum and Vicente 2015; Vicente 2018). In contrast, the coordination of different senses of a polysemous noun often results in oddness or ‘zeugma’. Consider:
(5) *The chicken is delicious and chirpy.
(6) *The newspaper fell of the table and fired its editor.
(1)-(6) are all copredicational sentences where the predicates in each case seem to select different senses of the noun. But while (1)-(4) seem syntactically similar to (5)-(6), the two sets of sentences clearly differ in felicitousness. What then explains the difference in felicitousness? With the simulation view of language processing in sight, I think the difference in felicitousness often has to do with the mental imagery we construct in processing these sentences.
Here’s a brief sketch of the account that I have been developing. On the simulation view, comprehending sentences frequently involves constructing mental imagery of a focal entity. A focal entity is usually indicated by the part of a sentence that grabs the addressee’s attention. For instance, in the sentence – ‘The red ball is on the table’ – the focal entity is plausibly ‘a red ball’. On the account I propose, zeugma in copredicational sentences often, though not always, arises when the conjunct occurring later in the sentence requires the language user to construct mental imagery of a focal entity that is different from the focal entity associated with processing the earlier conjunct.
Consider (5) ‘The chicken is delicious and chirpy’. One is likely to construct mental imagery of chicken-as-meat (e.g. mental imagery of a chicken fillet or chicken thigh) in processing the first conjunct, and that of chicken-as-animal (e.g. mental imagery of a live feathery chicken) in processing the second. In this case, the second conjunct seems to require the language user to construct mental imagery of a focal entity that is different from the focal entity associated with processing the first conjunct. In contrast to (5)-(6), comprehending (1)-(4) does not seem to involve such a conflict in focal entities, or if there is a conflict, it can be resolved during processing. Consider (1) ‘The book is thick and interesting’. Since our perceptual experience of book-as-content (the abstract realiser) is often just our experience of book-as-tome (the concrete realiser), the focal entity associated with processing the second conjunct would still be book-as-tome. In this case, the second predicate ‘interesting’ does not require the language user to construct mental imagery of a different focal entity. Similar explanations can also be given with respect to (2)-(4). Underlying this mental imagery account is the thought that given the syntax of copredicational sentences, language users, in trying to make sense of the sentence through simulation, expect there to be a single focal entity, and a clash in focal entities then generates zeugma.
Given what has been said here about the role of mental imagery in comprehending language, one might naturally wonder what all this means for aphantasics, who report not having conscious mental imagery. The phenomenon of aphantasia might seem to present a problem for the simulation view of language processing. It may well be, however, that aphantasics have unconscious mental imagery that is not available for introspective reports (Nanay 2021). While mental imagery involved in comprehending novel and poetic language is often conscious, that involved in comprehending ordinary, literal language need not always be consciously accessible. But questions about language comprehension with respect to aphantasics remain: What linguistic intuitions do aphantasics have about different copredicational sentences? Do they have difficulty appreciating poetic language that is supposed to evoke mental imagery? Do they have difficulty comprehending imagery-demanding metaphors? Like the research on mental imagery in relation to language comprehension, these questions remain to be investigated.
References
Asher, N. (2011) Lexical Meaning in Context, Cambridge: Cambridge University Press.
Bergen, B. (2015) Embodiment, Simulation and Meaning, in N. Riemer (ed.) The Routledge Handbook of Semantics, Oxford: Routledge, pp.156–173.
Bergen, B. & Wheeler, K. (2010) Grammatical Aspect and Mental Simulation, Brain & Language 112: 150-158.
Brunyé, T. T., Ditman, T., Mahoney, C. R., Augustyn, J. S., & Taylor, H. A. (2009) When You and I Share Perspectives: Pronouns Modulates Perspective-Taking During Narrative Comprehension, Psychological Science 20: 27-32.
Carston, R. (2010) Metaphor: Ad Hoc Concepts, Literal Meaning and Mental Images, Proceedings of the Aristotelian Society 110(3): 295-321.
Falkum, I. L. and Vicente, A. (2015) Polysemy: Current Perspectives and Approaches, Lingua, 157, pp.1–16.
Green, M. (2017) Imagery, Expression, and Metaphor, Philosophical Studies 174(1): 33-46.
Hauser, M. D., Chomsky, N. and Fitch, W. T. (2002) The Faculty of Language: What is it, Who Has it, and How Did it Evolve? Science, 298, pp.5598–1569.
Liu, M. (2022) Mental Imagery and Polysemy Processing, Journal of Consciousness Studies 29(5-6): 176-189.
Liu, M. (forthcoming) Mental Imagery and Poetry, Journal of Aesthetics and Art Criticism.
Nanay, B. (2021) Unconscious Mental Imagery, Philosophical Transactions of the Royal Society B, 376: 20190689.
Pustejovsky, J. (1995) The Generative Lexicon, Cambridge, MA: MIT Press.
Richter, T. and Zwaan, R. A. (2009) Processing of Color Words Activates Color Representations, Cognition 111: 383–389.
Stanfield, R. A. and Zwaan, R. A. (2001) The Effect of Implied Orientation Derived from Verbal Context on Picture Recognition, Psychological Science, 12(2), 153–156.
Vicente, A. (2018) Polysemy and Word Meaning: An Account of Lexical Meaning for Different Kinds of Content Words, Philosophical Studies, 175, pp.947–968.
Zwaan, R. A., Stanfield, R. A. and Yaxley, R. H. (2002) Language Comprehenders Mentally Represent the Shapes of Objects, Psychological Science, 13(2), pp.168–171.
Zwaan, R. A. and Yaxley, R. H. (2003) Spatial Iconicity Affects Semantic Relatedness Judgments, Psychonomic Bulletin & Review, 10(4), pp. 954–958.
Zwaan, R. A., Madden, C. J., Yaxley, R. H., and Aveyard, M. E. (2004) Moving Words: Dynamic Representations in Language Comprehension, Cognitive Science, 28, pp. 611–619.