Hold Your Definition

Blogging after quite a hiatus, more of which in the next post, and reading a first book since October, the two not unconnected.

Received Dan Dennett’s “Intuition Pumps – and Other Tools for Thinking” as a Christmas present. No secret here on Psybertron that I’m a big fan of Dennett, and Intuition Pumps is a retrospective reflection on many of his meta-thought-experiments about thinking, collected from his previous 45 years of writings. Many re-writes of pieces he’s published and presented several times himself and many, as he points out, anthologised multiple times by other editors. So in a sense, nothing new.

But Dennett’s voice is always readable and what this compilation brings is the selection and editorial commenting and re-phrasing, a cleaner re-phrasing of the core points stripped of any potentially misleading clutter. Dennett himself, as well as his reader, has learned a lot in 45 years. Even then, after 8 sets of 70-odd one-tool-per-chapter over 400-odd pages, there’s a whole chapter on what got left out (and where to find them). Not-included include the famous Where Am I examples derived from the Brain in a Vat thought experiment, nor the eight examples known as Quining Qualia.

One thing the editorial revisit brings, is rephrasing that counters any misleading interpretations introduced by earlier wise-cracking zingers intended to demolish adversaries. With hindsight, rhetorical put-downs may have overstated one’s argument and missed important lessons. The one example that pleased me most, given that I share Dennett’s belief that computer systems modelling does still and will continue to bring a great deal to the philosophy of mind and the brain-mind-consciousness problem, is the backtracking on the homunculus-as-infinite-regress view. The regress is of course finite, if each “controller” is a system of less intelligent controllers than the previous level, eventually the substrate really does comprise the dumb building blocks of chemistry and physics.

Intentionality and the intentional stance feature prominently of course, as does evolution as algorithm. The whole engineering take on evolution as problem solving – ladders, cranes and sky-hooks, scaffolding and staging, etc – gets an outing, whilst the many-layered properties of computer architectures maybe represents the single greatest part of the material.

In fact, I reckon chapter 24 on “Register Assembly Programming” should be compulsory education for all early secondary schoolers (7th/8th graders) independent of specific subject teaching. I vividly recall Hester (Mr Pearson) our Maths teacher recently converted from French teacher, running exactly the same pupils and boxes of beans exercise in class (though the beans may have been bits of paper IIRC). At the time I assumed we were learning how new-fangled computers worked (around 1971 this would have been) but what Dennett does is bring out and list explicitly the “Seven Secrets of Computer Power” – lessons of computing, not rules about computing, but rules about the world in general. [Post Note : Added my summary of Dennett’s rules here. And a later important reference here. And already linked above – the educational resource page.]

[Aside – Listening to Angie Hobbs on BBC R4 Saturday Live – wholeheartedly agree that philosophy needs to be taught in primary schools too, but the fascination with the analytical and rhetorical demolition tricks of paradoxes and pointless pre-socratic arguments must be supplemented with the tools of constructive solutions too. Otherwise philosophy remains the caricature counting angels on the head of a pin. From the mouths of babes – what do philosophers do? – pointless arguments about nothing all day long.]

If I may paraphrase that and the two subsequent chapters on algorithms and virtual machines – competencies in this layer as systems or patterns in an underlying layer of repeatable parts: Turing (and von Neumann and Church) already said it all; all advances since are about speed and power, not about any new rules or mechanisms; vis Rule 7, there are no new rules beyond Rule 6, full stop, end of.More complex layers are simply built upon less complex underlying layers, ad infinitum so far as necessary, and the whole is substrate neutral, any physics will do.

As I said none of this is entirely new to me, but one aspect I’d never seen explicitly, but maybe I absorbed osmotically from engaging with Dennett, was my own aversion to definitions. Throughout the book so far he uses the “sorta” operator, to allow approximate ontological definition within taxonomies. (Throughout this blog I use “kinda”.) Sure most of consciousness and intelligence, in fact most aspects of even dumb development through branching decisions – evolution itself – depends on the competency of making distinctions, detecting and acting on significant difference. But that doesn’t depend on tight definitions of those distinctions. It depends on their existence and significance. Like species, distinctions are defined with hindsight only. Rationality is 20:20 hindsight. So, as a philosopher who has straddled the border with science, with strong scientific sympathies (eg as one of the 4 horsemen in the science vs religion wars) Dennett remains resolutely a philosopher; even a whole chapter on that – why be a philosopher?

“One of [Dennett’s] guilty pleasures is watching eminent scientists, who only a few years ago expressed withering contempt for philosophy, stumble embarrassingly in their own efforts to set the world straight […] with a few briskly argued extrapolations from their own scientific research. Even better is when they request, and acknowledge, a little help from us philosophers.”

“There is no such thing as philosophy-free science, just science conducted without consideration of its underlying philosophical assumptions.”

“We should quell our desire to draw lines. We don’t need to draw lines. [Distinctions yes, but not with tight definitions.]”

“The intentional stance is the strategy of interpreting the behaviour of an entity by treating it as if it were a rational agent …. [with sorta consciousness, sorta intelligence, sorta beliefs & sorta aims in sorta life, etc.] ….

Define your terms sir! No, I won’t, that would be premature ….

Many philosophers cannot work that way; they [believe they] need utterly fixed boundaries to their problems and possible solutions ….”

A recommended read, whether you’ve read Dennett before or not.
A sorta greatest hits, selected by the author himself.

Post Note: Nice to see this Dennett extract quoted by Maria Popova at Brain Pickings, on the event of Dan’s birthday 28th March 2014. (Dennett credits these as Rappaport’s rules, but this is the version Dennett presents in Intuition Pumps.)

How to compose a successful critical commentary:

  • You should attempt to re-express your target’s position so clearly, vividly, and fairly that your target says, “Thanks, I wish I’d thought of putting it that way.
  • You should list any points of agreement (especially if they are not matters of general or widespread agreement).
  • You should mention anything you have learned from your target.
  • Only then are you permitted to say so much as a word of rebuttal or criticism.

These the first three rules I call elsewhere:

“Respect, respect & respect”.