notes: conceptual accessibility, and a formal approach to metaphorical analysis

metaphor

too much time and effort is spent arguing the merits and pitfalls of individual metaphors, when the root cause – a failure to adequately define and formalise ‘metaphorical-analysis’ – remains unaddressed

Undoubtedly, metaphor is innate to human experience, and cognition.

‘There is something that it is’ to intuitively perceive and consider universal phenomena by metaphor – to consider one kind of universal phenomenon in terms of the characteristics of another almost-entirely different kind of phenomenon.

–how can that be?

Equally, ’there is also something that it is’ to insufficiently define :

  1. What it is to perceive and consider by metaphor – what it means to do so, and perplexingly;
  2. How to set about it, in ‘as objective and volitional manner as possible’

The result, is that almost any conversational use of metaphor (particularly conversations where explicit specificity is valued: scientific discourse, for example) can quickly become distracted, then immersed, in ’the wrong stuff’

as with the metaphor of ‘cognition and computation’

actualmetaphormetaphormetaphorthingthingnot the actual metaphorcommon subset of concernsmetaphorthingset-of-all concernsset-of-all concerns

figure 1: metaphor

—what might be done?

metaphorical analysis

—what do we even mean?

Consider some phenomena B, as metaphor C.

When we consider phenomena B, as metaphor C :

  1. We aren’t saying B is literally C; or that B and C are absolutely the same
  2. We are saying some part of B and C is the same; that B and C relate by some partial, fractional equivalence

Given two sets-of-all circumstances: some elements of both B and C are included – are equivalent; and some elements of B and C are not.

CBset-of-all concernsset-of-all concernsCBset-of-all concernsset-of-all concernsCBACBACBsubset-of-common concernsCB

figure 2: metaphorical analysis

The above case and figure makes clear – we don’t need to know which elements specifically are included in our metaphor, to recognise that some elements aren’t - so arguing against a metaphor by the fact some elements don’t apply, is problematic; we ought to expect it.

—how do we minimise the chance of focusing on ’the wrong stuff’?

—how might we define which elements are included?

example: computation and cognition – what is a computer?

on seeing past distractions

What is a computer?

  1. The origin of the term was a person who performed computations
  2. We commonly use the term to refer to electronic silicon devices, which also perform computations
  3. $\ldots$

Consider 1 (a person who performed computations) and 2 (electronic silicon devices, which perform computations) as two special-cases of computer (noting that: there are more kinds, and we might expect more in the future)

Consider each (1, 2) as a set-theoretic set-of-all-respective-circumstances (B, C: figure 4), with intersect (A: figure 4). The intersection includes all that is common – the relative general-case – between both special-cases.

remember: we don’t need to know what is common, to know that some elements aren’t common; so we ought to be cautious when framing arguments by elements which aren’t {possibly; probably; demonstrably} common, at least

  1. This kind of (set-theoretic) analysis is such that the intersect between two arbitrary special-cases is always smaller than both, individually ($[\;|A| < |B|\;], [\;|A| < |C|\;]$: figure 4)
  2. Further, if we imagine a hypothetical set-of-all computers, whereby each computer is represented by a set-of-all respective circumstances (of each special-case), we might rightly imagine the intersect (E: figure 4) between all – the ’essential general-case’ – is smaller still ($|E| < |A|$: figure 4); its’ elements simpler, and more abstract, than the material implementation of {each; any} special-case
name

figure 4: essential general case

This ’essential general-case’ – the resultant set of the intersect between all kinds of computer (with simpler, more abstract elements than any material implementation individually) – is a good approximation for our computer metaphor (sufficiently-equivalent for this example), and importantly, more explicitly frames the importance of ignoring special-case details from the metaphor itself.

for more see: on cognition and computation