Notes — Steven Pinker on Language, Reason, and the Future of Violence

Source: Conversations with Tyler, Episode 14, 2 November 2016. Live conversation at George Mason University.

Four questions [Adler frame]

Q1 — What is it about? A wide-ranging conversation spanning Pinker’s career: the dual-mechanism account of irregular verbs, the limits of Chomsky’s universal grammar, the hard problem of consciousness, the case for reason and progress, and a preview of Enlightenment Now. The underlying thread is that human cognition and human history are both more structured and more tractable than blank-slate or irrationalist views suggest.

Q2 — How is it argued? Through worked examples at every level. The irregular verb is used to show two mechanisms competing; Hume is invoked to distinguish the conceptual claim (reason can’t derive goals) from the psychological claim (people are permanently irrational); the Fermi paradox is addressed by challenging the premise that advancing civilisations converge on self-destruction. The method is: dissolve the dichotomy, go to the level of mechanism, and let the evidence constrain the theory.

Q3 — Is it true? The dual-mechanism account of language is well-supported empirically. The mysterianism position on consciousness is defensible but contested — Dennett would say the ‘hard problem’ is a pseudo-problem; Lennox and McGilchrist would say the problem points beyond materialism. The Enlightenment progress argument is empirically grounded for the dimensions Pinker quantifies (violence, poverty, health) but the causal attribution to Enlightenment institutions is contested. Group selection rejection is mainstream in evolutionary biology.

Q4 — What of it? The dual-mechanism view has practical implications for language pedagogy and style (preserve the passive voice where logically warranted). The progress argument provides a counterweight to apocalypticism. The mysterianism view blocks quick metaphysical moves in either direction — neither the materialist dismissal of consciousness as illusion nor the theist move to a non-physical soul.

Glossary

Irregular verb: a verb whose past tense is not formed by adding -ed to the stem (e.g., ‘sang’, ‘went’, ‘came’). In Pinker’s account, these are stored as individual lexical entries in the memory system, not generated by rule.

Dual-mechanism model of language: the theory that language rests on two distinct cognitive systems — a memory system that stores arbitrary pairings (words, irregular forms) and a rule system that generates forms compositionally (grammar, regular -ed past tense). The systems can compete over the same communicative function.

Universal grammar (Chomsky): the hypothesis that the human mind comes with an innate, language-specific learning device that biases children towards certain types of grammatical rules. Contested at the level of what exactly is built in; less contested as the claim that the mind brings structure to language learning.

Mysterianism: the position that the hard problem of consciousness — why physical processes feel like something — may be permanently beyond human intuitive grasp, not because of undiscovered physics but because the mind reflecting on itself is constitutionally limited. Associated with Colin McGinn; anticipated by Tom Nagel.

Hard problem of consciousness: David Chalmers’s term for the explanatory gap between complete physical/functional description of a brain and the fact that there is subjective experience at all. Contrasted with ‘easy problems’ (explaining the functions of perception, attention, memory).

EAAN (Evolutionary Argument Against Naturalism): not directly named by Pinker but relevant — if the mind is purely the product of blind evolution, there is a question whether its faculties are truth-tracking. Pinker dismisses the premises: we know from neuroscience that there is no aspect of consciousness without a physical correlate.

Euphemism treadmill: Pinker’s term for the cycle in which a neutral word for a stigmatised referent gradually acquires negative connotations (because the connotation attaches to the referent, not the word), prompting replacement by a new neutral term, which in turn acquires the same connotation.

Group selection: the hypothesis that natural selection operates on groups as well as individuals and genes. Pinker rejects it as biologically incoherent; the relevant unit of selection is the gene, not the group.

Section notes

Dual mechanism: memory and algorithm

The irregular-verb research is philosophically productive because it isolates two cognitive mechanisms at work in the same linguistic domain, competing for the same function. The key finding: irregular verb frequency predicts survival. ‘Snuck’ survived because it could parasitise the phonological pattern of ‘struck’, ‘stuck’, ‘stunk’. The Hayekian reading (Cowen’s framing) is that frequency selection is a decentralised, unplanned process — the language as a whole evolves toward some balance without any agent designing it.

Pinker prefers Darwin to Hayek: the relevant mechanism is Darwinian filtration through individual memories, not market-like adjustment.

Universal grammar: what remains

Pinker’s measured verdict: there is an innate structure that does the learning. This is not ‘the grammar is fully innate’ (Chomsky’s earlier position) nor ‘there is no innate linguistic structure’ (empiricist). The claim is computational: the child brings a learning mechanism with certain biases, not a full grammar.

The question of how precise those biases are is left open. The point for Pinker is that this dissolves the nature-nurture dichotomy: what is innate is an ability to learn, and the nature of that ability is the substantive question.

Consciousness: the wall, not the gap

Pinker’s position is neither eliminativist (Dennett) nor dualist. He acknowledges that:

• All conscious experience has a physical correlate
• This means no extra-physical causation (no ESP, no action at a distance)
• But why any physical process feels like something is not answered by any of this

He invokes Colin McGinn’s mysterianism and Tom Nagel’s ‘What Is It Like to Be a Bat?’ The key move: ‘there are some facts about the universe that are true, and we’ll never be satisfied that we intuitively understand them.’ The parallel: what could have been before the Big Bang? How can the universe be either finite or infinite? Some things are cognitively closed.

Comparison with Lennox [§ cross-source]: Lennox uses the same ‘hard problem’ to argue that materialism cannot explain consciousness, and therefore that there must be a non-physical dimension to reality. Pinker denies the inference: the fact that consciousness is puzzling does not mean it is non-physical. The hard problem is a wall, not a window.

Comparison with McGilchrist: McGilchrist similarly holds that the reductive account of consciousness misses something essential, but grounds this in the different modes of attention of the two hemispheres rather than in mysterianism. Both Pinker and McGilchrist resist easy eliminativism; they diverge on what the resistance implies.

Reason and the Enlightenment (in preview)

The 2016 conversation is the first public articulation of the Enlightenment Now thesis. Key claims:

1. By the very act of questioning reason, you are committed to reason.
2. Hume’s point that reason is a slave to the passions is conceptual (you can’t derive values from logic), not psychological (people are permanently irrational).
3. Progress is empirically measurable: declining violence, poverty, disease.
4. Progress is explicable as the fruit of Enlightenment institutions (science, liberal democracy, markets).

Cowen pushes back: he is more Hayekian, more sceptical of reason’s powers. Pinker’s reply: the frontal cortex allows periodic override of passion; this is sufficient for the Enlightenment case; the evidence is the historical record.

Group selection (Q&A)

Pinker rejects group selection as a Darwinian mechanism in the biological sense. The timescale argument is decisive: Germany went from the world’s most militarist culture to the world’s most pacifist in roughly one generation. Gene-frequency change cannot operate on that timescale. The mechanisms are cultural and institutional — changes in norms, laws, and narrative, not changes in the gene pool.

[?] Flags

• [?] Pinker mentions ‘mycoticism’ in the context of historical momentum — this appears to be his term for treating historical trends as having the properties of physical inertia. Worth checking against Enlightenment Now chapter structure.