Time Heals All Wounds, And Then Kills the Patient
<Previous Next>
Tue Apr 1 05:14:02 2014
Typing in Colours

(Cross-posted to G+, but it's more of a definitive statement of views so it goes here too)

A recent instance of 「Wasted Talent」: here

I'm not normally the kind of person to post links to episodes of webcomics, but maybe I can spin this one into being about philosophical topics. It coincides with a recent XKCD(note 1): https://xkcd.com/1349/ , and together we can use them to approach three related topics: abstraction, framing, and deconstruction. In these two comics, the humour comes from choosing to look at a task (programming) from a very unusual frame, in particular one that's much more low-level than the one we normally use; the "typing in colours" and the "working with sand" are akin to a common flight of fancy from desperate programmers struggling against a deadline - we often quip that our job is to push bits around, and if only one could know some right sequence of keys to push that would generate a program (or program component) that meets the spec, one might meet a deadline that's looking like it's going to slip because .. well, we need to think and test a lot to actually work out the right code structure we need. Even though it's true that what divides us from being done is "just bits", what divides any person from being a great (and rich) novelist is "just words", or any person from being a great musical composer is "just notes"; the creative act is not that easy even if the physical motions are.

So, the terms. Abstraction is the layering of conceptual levels at which we might work with something, with the lower levels being a "zoomed in" view, the higher levels being a "big picture". Looking at humans at a high level of abstraction might have us seeing societies and their traits, at a lower level subcultures, at a lower level individual people, at a lower level organs, at a lower level yet cells, and at a lower level molecules, then atoms, and so on. In theory, most physical things have a common origin if you look low enough (although, interesting for epistemologically-focused people, we have a practical-world-outlook system where we don't know the root(note 1)). The further we get from reasonably physical areas of our abstractions from the universe, the more we're projecting meaning onto the universe (and treating things as pattern-object hybrids) ; does it make sense, for example, to include subcultures in our model? The question of whether conceptual objects in higher abstraction levels actually exist becomes fuzzy - not even necessarily empirically fuzzy, but philosophically fuzzy(note 3)

Framing is the act of wrapping around a set of relatively bare facts a set of ideas - metaphors, historical context, stories, other information, in order to suggest structure for (dealing with/thinking about) those bare facts. We may work with one framing (and this is easy to do; framing relieves the stress of not knowing how to think about something), or we may develop habits where we try to find multiple framings for the same relatively-bare-facts and compare/contrast them.

Deconstruction is a discourse-technique(note 4) in which one takes an existing framing, in whatever particular form it is exposed (a narrative? even a word?) and examines its roots enough that its hold is weakened; this works best on framings that have strong(note 5) competition or weak construction(note 5).

Abstraction and Framing have a complex relationship. One classic example is Margaret Thatcher's statement: "they are casting their problems on society and who is society? There is no such thing! There are individual men and women and there are families and no government can do anything except through people". Here Thatcher offers an argument(note 7) to deconstruct the notion of society, noting that is composed of individuals and asking us to give up on the higher-level of abstraction because the lower level is adequate. It is a kind of "there are no forests, there are only trees" type statement, but it has a certain force to it (note that I am not sympathetic to the statement, but it is a good example of this). Not all framings are physical-chain abstractions - in fact, only a few are. Looking at an attack as an isolated event or as part of a historical chain that involves recent other attacks the other way is a type of framing. Although we might decide to build abstractions higher up from a chain of events, building some conceptual object that encapsulates features we find common and salient about them, and possibly dividing them from similar-ish events that have a few differences. "Freedom fighter" versus "terrorist", perhaps? "Justified response" versus "atrocity"?

One of the features of society is that we have a seeming coherence between people in the concepts and language and framing we use to understand the world. We can communicate, usually, and we don't seem mentally that alien to each other; we trust that the fact that we're living in the same universe and seem to be speaking the same language makes sharing meaning between our mental world to be easy. One of the great difficulties between people is that this is not particularly true. There are many degrees of freedom in the concepts and frames we use to parse the world, and the people who think deeply about these things(note 8) rarely enough converse with enough depth or discipline to expose the deep differences that are inevitable when people build meaning alone (or in subsets of society). Some people have enough skill with language and navigating differences in meaning to actively shape framing - debaters and philosophers both have this as their craft (with different emphases). Still, we expect the world to be full of people who haven't reflected enough on their own world-of-terms and concepts to really understand them(note 9), with innumerable differences between people lurking just beyond the surface - the differences are usually masked by language not being high-enough-bandwidth to expose the gaps, but the differences are also not usually so large that the basic ability to communicate, particularly on mundane/functional matters(note 10), has too low an actual conceptual bandwidth(note 11).