# Against Perceiving Meanings: Empedocles Leaped or Loves?  ## Metadata - Author: [[245]] - Full Title: Against Perceiving Meanings: Empedocles Leaped or Loves? - Category: #articles - Summary: insert summary - My notes: - Summary: The strong hearing-meanings view says we hear sentence meanings in our auditory experience. When combined with models of bilingual speech, this view implies that beliefs heavily shape what we hear. This leads to a problem because it requires constant, widespread cognitive influence on perception, which is not well supported by evidence. - URL: https://readwise.io/reader/document_raw_content/408040351 ## LLM Chats ## NotebookLM ## LLM Audio ## Highlights > In reply, Brogaard (2017, 2018, 2020) argues that there *is* a phenomenal difference between hearing such homophones in context, and that this supports a strong version of the hearingmeanings thesis. The disagreement between O'Callaghan and Brogaard appears to rest on conflicting introspective judgments about what these experiences are like. On the face of it, then, the dialectic threatens to collapse into a stand-off over the reliability of introspection. ([View Highlight](https://read.readwise.io/read/01khjjgbj471cy5eh1fhk6v4zn)) > Empirical work suggests that, in fluent bilinguals, lexical candidates from both languages are activated non-selectively, while extra-linguistic cues, encoded in beliefs and expectations about which language is being used, guide the inhibition of non-target meanings. ([View Highlight](https://read.readwise.io/read/01khjhyxjg9d7n6865txzpn7gg)) > For bilingual speakers who are fluent in both languages, which meaning is understood in such cases is sensitive to extra-linguistic information about which language is being used—for instance, information about the conversational setting or about the interlocutor's linguistic profile. ([View Highlight](https://read.readwise.io/read/01kj2fmmp7dsa4x8bqat22jvxn)) > I argue that, if a strong version of the hearing-meanings thesis is applied to such bilingual homophone cases—one on which distinct sentence meanings are themselves part of auditory phenomenal character—then the resulting view incurs a substantial commitment to pervasive, belief-driven cognitive penetration of perception. I suggest that this commitment is both theoretically costly, given standard ways of drawing the perception–cognition boundary, and poorly supported by the empirical literature on speech perception and bilingual lexical access. ([View Highlight](https://read.readwise.io/read/01kj2fnfkrsg4erx9bs91tadw7)) > Methodologically, the proposal also illustrates how > independently motivated psycholinguistic models of bilingual comprehension can constrain otherwise purely phenomenological arguments in the philosophy of perception. ([View Highlight](https://read.readwise.io/read/01kj2fq328bx16dvysb9pb85sg)) > Second, the thesis concerns the resulting state rather than the processing route. It is compatible with meaning comprehension involving various cognitive and inferential processes, provided that these processes terminate in a perceptual experience whose phenomenal character is partly fixed by the sentence's meaning. In this respect, high-levelism about meanings is distinct from non-inferentialism or "perceptualism" about speech comprehension, which focus on whether the process is inferential. ([View Highlight](https://read.readwise.io/read/01kj2g0wm4grd6p3c38evb4mb4)) > Third, the thesis is explicitly about phenomenal character. A view on which meanings figure in the representational content of perceptual states, or are detected by perceptual mechanisms in the brain, but make no distinctive contribution to what it is like to listen to speech, would not count as a high-levelist view in the present sense. ([View Highlight](https://read.readwise.io/read/01kj2g10zhmf2hc8y1jbtefa7b)) > At this point, the dispute seems to turn largely on whose introspective verdict one finds more credible. Brogaard and O'Callaghan simply disagree about what it is like to hear these sentences, and there is no independent way, within this dialectical frame, to adjudicate between their reports. More generally, the force of phenomenal contrast arguments depends on a sufficient degree of agreement about the relevant phenomenal similarities and differences; when such agreement breaks down, introspective methods appear to reach their limits (cf. Block 2014; Burge 2014; Chudnoff 2015; Koksvik 2015). ([View Highlight](https://read.readwise.io/read/01kj2ge9w76d9whrbb50gthbpk)) > In each pair, a single acoustic pattern can serve as a well-formed utterance in either language, yet the propositions expressed are entirely different. ([View Highlight](https://read.readwise.io/read/01kj2gsd6m9a0msze72nvpbkvp))