Chapter 7: Nonliteral Language Processing

Key Focus: Types of Nonliteral Language, Metaphor Theories, & Neural Bases

📖 Required Reading

• Chapter 7 from Introduction to Psycholinguistics: Understanding Language Science (2nd ed.) by Matthew Traxler.
  Link to Chapter 7

Overview

This chapter examines how we process nonliteral language - expressions where literal meaning differs from intended meaning. We explore various types of nonliteral language (indirect requests, idioms, metaphors, irony), compare major theories of metaphor comprehension, and investigate the neural basis of nonliteral processing. The chapter explains why we use nonliteral expressions and how they enhance social interaction and communication efficiency.

Learning Goals

After studying this chapter, you should be able to:

• Differentiate between major types of nonliteral language (indirect requests, idioms, metaphors, irony)
• Evaluate evidence against the Standard Pragmatic View of literal-first processing
• Compare competing theories of metaphor comprehension (Comparison, Property-Matching, Class-Inclusion, Career-of-Metaphor)
• Explain the social, communicative, and cognitive benefits of using metaphors
• Distinguish between metaphor and metonymy and explain how metonymy is processed
• Understand how embodied cognition contributes to nonliteral language comprehension
• Analyze the neural basis of nonliteral processing and the Graded Salience Hypothesis
• Apply nonliteral language principles to explain real-world communication phenomena

1. Introduction to Nonliteral Language

1.1 Core Definition

Nonliteral language = when literal meaning ≠ intended meaning.

Foundational Examples

1.2 Key Challenge: The “Recognition Problem”

1. Recognition: How do we know to ignore literal meaning? (e.g., Can you open the door? = request, not ability question).
2. Computation: How do we derive the intended meaning once nonliteral intent is recognized?

2. Types of Nonliteral Language

2.1 Indirect vs. Direct Requests

• Direct: Explicit commands (e.g., Open the door!, Tell me the time!).
• Indirect: Polite, masked as questions/statements (e.g., Do you have the time? = “Tell me the time”; Would you pass the salt? = “Give me the salt”).
  • Why indirect? Reduces “face threat”—rejection is less awkward (e.g., “No, I can’t reach” vs. “No, I won’t”).

2.2 Idioms

Fixed phrases where whole meaning ≠ sum of parts:

• Non-decomposable: No link between words and meaning (e.g., kicked the bucket = “died”; spilled the beans = “revealed a secret”).
• Decomposable: Words map to meaning components (e.g., pop the question = “propose marriage”; “pop” = “ask suddenly,” “question” = “proposal”).

2.3 Metaphors

Comparisons between unrelated concepts to convey abstract ideas:

• Attributive: A is B (e.g., That lecture was a sleeping pill = “The lecture was boring”).
• Relational: Highlight concept relationships (e.g., Doctor : Patient :: Teacher : Student).

2.4 Irony/Sarcasm

Saying the opposite of intent (context signals criticism/humor):

• Example: After a boring lecture → Now that was exciting! = “That was boring.”

2.5 The Standard Pragmatic View (Debunked)

Traditional Claim: Process literal meaning first → reject/revise if nonsensical.

Evidence Against:

1. Gibbs (1983): Indirect requests (Can you open the window?) processed as fast as direct requests—no “literal first” delay.
2. Ortony (1979): The sheep followed their leader (literal = animals; nonliteral = investors) had identical reading times—no extra effort for nonliteral.
3. Stroop Task (Glucksberg et al., 1982): Slower to reject Keith is a baby (literal false, metaphor true) than Keith is a basin (literal false)—nonliteral meanings activate automatically.

3. Metaphor Comprehension Theories

Metaphors = assign properties of a vehicle (describing concept) to a topic (focus).

3.1 Key Terms

• Topic: What the metaphor is about (e.g., Nicole Kidman in Nicole Kidman is bad medicine).
• Vehicle: Concept providing properties (e.g., bad medicine = “dangerous, harmful”).

3.2 Theory 1: Comparison View (Conversion to Simile)

• Claim: Metaphors are converted to similes (A is like B) to find shared properties.
  • Example: Nicole Kidman is bad medicine → Nicole Kidman is like bad medicine → “dangerous” is shared.
• Limitations:
  • Irreversible: My surgeon is a butcher (unskilled) ≠ My butcher is a surgeon (skilled)—similes are reversible.
  • Slow processing: Predicts metaphors take longer than similes, but experiments show metaphors are faster.

3.3 Theory 2: Property-Matching Hypothesis

• Claim: List vehicle properties → filter irrelevant ones → match to topic.
  • Example: bad medicine → “dangerous, makes sick, comes in bottles” → filter “comes in bottles” → “dangerous” → Nicole Kidman is dangerous.
• Subtheory: Salience Imbalance:
  • Literal similes (Copper is like tin): Shared properties are salient (obvious) for both.
  • Metaphors: Shared properties are salient in the vehicle, non-salient in the topic.
• Limitations: Fails for metaphors with low-salience shared properties (The senator was an old fox = “clever”—not obvious for either).

3.4 Theory 3: Class-Inclusion Hypothesis (Dominant)

• Claim: Metaphors = category membership statements—topic is in an ad hoc category (temporary, context-dependent) exemplified by the vehicle.
• Dual Reference: Vehicles have two meanings:
  1. Literal: Basic concept (shark = fish).
  2. Superordinate: Ad hoc category (shark = “dangerous, aggressive things”).
  • Example: My lawyer is a shark → Lawyer is in the category “dangerous things” (exemplified by sharks).
• Evidence:
  • Modifier compatibility: My lawyer is a well-paid shark (sensible—“well-paid” applies to the category, not literal sharks).
  • Priming: Priming literal sharks swim slows metaphor comprehension (draws focus from ad hoc category).

3.5 Theory 4: Career-of-Metaphor Hypothesis

• Claim: Metaphors evolve from novel to conventional—processing shifts:
  • Novel (Dancers are butterflies): Processed via comparison (find shared properties: “graceful”).
  • Conventional (Problems are roadblocks): Processed via class-inclusion (access ad hoc category: “things that block progress”).
• Evidence: Participants prefer similes for novel metaphors, category form for conventional ones.

4. Why Use Metaphors?

4.1 Social Politeness

Reduce awkwardness in sensitive contexts:

• Example: If you’re free Saturday, I know a great restaurant (instead of Let’s date)—rejection (“I’m busy”) is less awkward.

4.2 Communicative Efficiency

Pack complex meaning into concise phrases:

• Example: You can be my Yoko Ono → Conveys “accompany me, even if you irritate friends/disrupt work” (alludes to Yoko Ono’s role in the Beatles’ breakup).

4.3 Enhance Comprehension/Memory

Map abstract concepts to familiar domains:

• Experiment (Albritton et al., 1995):
  • Metaphor group: Read crime text with crime is disease (“diagnosed sources, sought cure”).
  • Literal group: Read with literal phrases (“studied sources, sought solution”).
• Result: Metaphor group recalled 20% more details—“disease” provided a coherent framework.

5. Metonymy & Under-Specification

5.1 What is Metonymy?

Metonymy = word refers to something associated with its literal referent (not similar, like metaphor).

Types of Metonymy

5.2 Processing Metonymy: Under-Specification Hypothesis

1. Initial Activation: Broad associated concepts activate (e.g., Dickens = person, books, legacy).
2. Homing-In: Context narrows to the right meaning:

• I read Dickens → “books” (context: “read”).
• Met my great-great-grandmother met Dickens → “person” (context: “met”).
• Evidence: Familiar metonymies (Protested after Vietnam) processed as fast as literal phrases (Visited Vietnam).

6. Embodiment & Nonliteral Language

6.1 Core Claim

Nonliteral comprehension relies on mental simulation of physical experiences (e.g., bodily sensations, actions). Abstract concepts map to concrete domains.

Example: Anger as Heated Fluid

• Idioms like blow your stack, flip your lid draw on embodied pressure/heat experiences.
• Evidence:
  • Reading John blew his stack speeds responses to heat (vs. chair).
  • Physical actions (e.g., “swallowing”) speed processing of idioms like swallowed his pride.

7. Neural Basis of Nonliteral Language

7.1 Hypothesis 1: Right Hemisphere (RH) Hypothesis (Debunked)

• Claim: RH = nonliteral processing; LH = literal processing.
• Counterevidence:
  • fMRI (Rapp et al., 2004): Metaphors activated LH (inferior frontal gyrus) more than RH.
  • Shibata et al. (2007): LH more active for metaphors; RH more active for literal sentences.

7.2 Hypothesis 2: Graded Salience Hypothesis (Giora, 2003)

• Claim: Hemisphere use depends on meaning salience (frequency/conventionality):
  • LH: Processes salient meanings (conventional metaphors, literal; e.g., paper tiger = “weak threat”).
  • RH: Processes non-salient meanings (novel metaphors; e.g., pearl tears = “precious tears”).
• Mechanism:
  • RH: “Coarse coding” (broad concept activation—good for novel links).
  • LH: “Fine coding” (narrow activation—good for conventional links).
• Evidence:
  • fMRI: Novel metaphors activate RH; conventional activate LH.
  • TMS: Zapping RH disrupts novel metaphors; zapping LH disrupts conventional ones.

Quick Review Questions

1. What is the difference between an indirect request and a direct request? Give an example of each.
2. How does the Class-Inclusion Hypothesis explain My lawyer is a shark?
3. Why are metaphors more efficient than literal language?
4. What is metonymy? Give an example of producer-for-product metonymy.
5. According to the Graded Salience Hypothesis, which hemisphere processes novel metaphors? Why?