A Typology of Abstractions


This is derived from previous stuff in the section on Questions of Resonance.

It's the first draft of a typology of abstractions

Introduction
The idea of abstraction, here, is based on patterns of use (rather than of 'recognition' or even 'cognition' per se) - so the starting point is fundamentally semiotic: i.e. Barthes's idea that "every use becomes [or can become] a sign of itself". The process of 'abstraction' is, fundamentally, a process of stripping away some uses to create new uses, often in new forms or articulations, rather than stripping away perceptions, so this is a kind of pragmatics of abstraction. Of course this also involves semiotics as a 'difference machine' - discovering and articulating differences between things and the actions associated with them, but these can all regarded as uses - of gestures, words, objects, processes, ideas, etc.

Or, to put it another way, and to make matters more interesting, this idea of the semiotics of use resonates with J. J. Gibson's ideas of 'perceptive-action' and affordances: i.e. the idea that the distinction between 'perception' and 'recognition' on the one hand, and 'use' or 'action' on the other hand is at least not very helpful, and at most just a (negative) legacy-distraction from pre-ecological (or 'primitive' - in the literal sense of the term) psychology. Gibson held that it's all one process (even though it is quite possible to separate 'perception' from 'action' analytically).

michelangelo awakening slave.jpg
Michelangelo: Awakening Slave (unfinished)
The example of science is a good place to start on abstractions:


Science starts with a process of 'stripping away' (see Iain McGilchrist's example of Michelangelo's unfinished sculpture, left), which is simultaneously a process of removing and decluttering, and a creative process of revealing. (A sculpture is a figure created by sculpting absences; it ends up 'enclosed by' absences, which were 'cut out' from the original block).

To return to physics ...
What happens, for instance, in the first uses of the word 'motion' in physics is that physicists (Newton, in particular) appropriated a word in everyday use ('motion') and then stripped and pared it down to the scientific term, 'motion', which no longer depends, nor is it linked, networked, associated, contaminated (etc) by any particular subject or context (person, agent, institution, culture).

In the process, the scientific term becomes 'pure [intellectual] currency' - stripped of subjectivity and context so thoroughly that anyone (or any agent), anywhere, can use it at anytime. This, of course, is an extrapolation of the ideal notion of monetary currency, which would (if it existed) similarly have 'universal' exchange value. (And the early forms of scientific proofs - Athenian geometry), emerged about the 6th Century B.C, at the same time as their currency - an early, very strong, cross-modal resonance).

So ... 'abstraction' in this particular sense - (there are several others, see below) - is paradoxically based on considerable narrowing down of a particular (scientific, meta-semiotic) use of the word 'motion', but at the same time massively broadening out the set of prospective use/rs, to include the universe of all possible agents (or actants, in Latour's sense), in all possible contexts. On top of that is of course a codicil, i.e. the falsifiability, or contingency of all scientific knowledge - the idea that scientific knowledge is only good for as long as it doesn't get refuted, or is even, to put a point on it, 'unreplicable'.

And to make matters even worse (or better, if you think it's more interesting) this process of 'universalised' scientific knowledge (e.g. the genome of anthrax) removes, or at the very least 'uncouples' this kind of knowledge from responsibility. Removing ties to context and subjects, de facto, removes responsibility. So this kind of abstraction is a process of creating (potentially) 'rogue' abstractions, and even non-responsible knowledge.

Which leaves me to think that the term 'abstraction' itself is a useful example of how abstractions (plural) are actually a 'resonant family' of tropes of stripping out, stripping away, revealing, and shifting above, shifting across (add to taste ...) ... relative to: modes, tropes, subjects, contexts, eras, 'paradigms' (add to taste here too ... ).

How does Abstraction relate to Knowledge?
Resonant knowledge/s covers a large range of types of knowledge, which are based on a range of levels and types of abstraction. So, it might be useful to unpack the range and types of abstraction...

We start with a working definition of knowledge as the capacity for effective action (add link ...), and we also start with the foundations of semiotics: the idea that "every use becomes a sign of itself", or perhaps more specifically: 'every use can become a sign of itself'. These definitions are grounded in pragmatics, although they very quickly develop in other directions too - epistemological, ontological, evolutionary psychology, etc, all of which are part of the broader field of socio-genesis.

In a very real sense, as signs become more complicated, and more abstract, they 'lift off' from their material, sensory, and pragmatic origins. (This is similar to Marx's analysis of the way exchange value is 'lifted off' from use value) (add link), and it also links to Rossi-Landi's definition of 'work' as 'appetite held in check', (add link) or, as other writers might say, 'deferred desire', placing time (and history), and relationships at the foundations of semiotics.

Abstractions can be good and bad news, as they provide stronger, more deft, higher resolution, more exchangeable (etc) semiotic machines, systems of difference, and algorithms and code - for synthesis into further abstractions and programmes - all of which, in plain language are, simply, tools for (more powerful) description and use. But they can, by definition, also get into the 'wrong hands'. The dynamics of language and (other projections of) power are well illustrated in Wopko Jensma's two drawings (1975) ...
Jensma 3
"I can dominate you if I know more words than you do".
Jensma 4
Mr Highbrow meets Mr Lowbrow

















Types of abstraction

Working (or playing) with these notions, we can identify about 11 different types of abstraction (plus two 'subtractions'), although they overlap to some extent.

-1. Questioning Capitalism?
As we approach 2020, it seems that we might, first, need to revisit the working definition of abstraction (see above), and apply it to the abstractions of capitalism, to outline some of the limits of this analysis, and the limits that we seem to be hitting of our own practice:
  • The process of 'abstraction' is, fundamentally, a process of stripping away some uses to create new uses, often in new forms or articulations, (which creating new ways of seeing) rather than stripping away perceptions per se, so this is a kind of pragmatics of abstraction.

The practice of capitalism starts with the establishment and expansion of mercantilism, then bearer-bonds, capital markets, capital futures markets, then derivatives of derivatives of derivatives (2008-style), etc, which become hyper-commodities and playthings in the hands of speculators, billionaires, oligarchs, and bankrupts.

As we move into 2017, it has become fashionable to question 'capitalism', &/or 'globalism' (see Chris Anderson of TED, here), but that's missing the point. Markets are based on the semiotic metamorphosis of use value into exchange value (as Marx pointed out so long ago), which takes on a life of its own, increasing its distance from, and alienation from, use value, as the derivatives spiral out of contro. This is perhaps the most fundamental semiotic metamorphosis of social life. The process is in principle and in practice reversible, and markets are in principle complex-adaptive systems (i.e. large numbers of agents interacting and communicating frequently, with fairly large degrees of freedom / limited ('negative') constraints, in ways that are ordered, but not, ultimately, predictable, and no agent can ever know what is going on in the whole system). Markets should, within these criteria, be capable of self-organisation and self-regulation.

However, that depends entirely on the boundaries, the negative constraints (or lack of constraints) that are in place, and agreement on the agency and siting of the power to exercise those constraints.

Command economies, like the Soviet Union, tried total regulation of markets, but found out, after many decades, that that's a contraction in terms. On the other hand capitalism requires mechanisms for the accumulation of capital, and its redeployment in areas of innovation. But without constraints, this can result in obscene accumulations of capital, and at least grossly indecent discrepancies of expenditure and consumption.

The key to these perversions of the markets is the creation of hyper-spaces or virtual-spaces, which are beyond the constraints of representative democracy (or the remnants of 'command' economies, for that matter). These hyperspaces do not even have to be physically 'off' shore - (e.g the State of Delaware in the USA [and some say even London] is effectively 'off-shore') - and to the extent that they are regulated, they are self-regulated by the global kleptocracy itself.

This neatly divides global markets into two exclusive zones - the 'national' ones, in which 'democratic' institutions seek to constrain budgets through fiscal and monetary 'controls' - to try to ensure reasonable distribution of wealth and opportunity, and the hyper-national ones in virtual-space, which the global kleptocracy controls, and seeks only to accumulate more and more capital, and influence: a sphere of control which is actively aided and abetted by most of the leaders of the national states themselves, most of whom have financial stakes in these virtual spaces. To make it even more interesting, 'sovereign wealth' funds are, surely, now investing in these hyper-national companies/'financial instruments' too. (This is of course nothing new, but it has become starkly more evident in recent decades).

The supreme irony is that it is only the 'nationalist' (or in 2016/7 the 'populist'), 'independent' surges within 'sovereign' nations that prevents any effective global agreement being reached on the regulation of these global hyper-markets. This kind of 'populism' (of populist leaders, not the populace), and lack of global agreement, guarantees that the worst aspects of globalism continue. Prime Minister Gordon Brown tried to negotiate agreement on global regulation post-2008, but was soon dismissed. Trump, with many (or most) of his companies registered in Delaware (see above), is playing this game from both ends. He may look inept, but that doesn't, necessarily, mean he's stupid. Fiercely independent 'sovereign' nations are the best guarantees that the global kleptocracy's wealth and power will always be assured, and protected from meddling nation states.

States that can forge a resonance, and a unity of purpose amongst/across themselves to achieve some measure of equity are possibly the building blocks for eventually constraining the global kleptocracy. The European Union was a nice try, but it needs radical reform - as seen its recent 'free trade' negotiations with Canada and the USA, which threatened to further emasculate the democratic process in favour of a free for all - for the transnational hyper-markets.


0. Subtractions

In the 24/8/16 post, here, I asked whether there was a relationship between abstraction and subtraction, and if so, how could we describe it? I also, in the 30/8/16 post, here, raised the question of how this project on resonance and knowledge could deal with the question of complexity, or more fully for the purposes of this project, complex adaptive systems. Following John Harris's seriously unhelpful article, here, I'm afraid there is nothing for it but to headline it in this section, '0', and get it out of the way, now.

So ...

Two issues of subtraction, both often masquerading as abstraction, need to be dealt with here : paternalistic reductionism, and regressive emotionalism, both of which tend to spill over into elitist or populist tribalism / 'nationalism'.

The media are currently collaborating in the attempt to normalise and domesticate emotionalism (aka 'populism'), and present it as the key to understanding the realpolitik of the new, 'post-truth' society. I prefer the anti-cerebral definition of an anti-truth society: "I feel therefore I am", see the post of 30/9/16, here.

Emotionalism is much more than just 'anti-intellectualism', or as Michael Gove would have it, 'anti-expertism'. It's not just about how 'they' feel, and act, it's about all of us. It's anti-corticalism - a regression to an evolutionary stage prior to the development of the cerebral cortex. This is, of course, not to say that emotions are not key to our makeup as human beings. Far from it. On the contrary, one of the advantages of social media (yes, there are advantages) is that (some of) the new networks incorporate ideas and emotions, and no longer subscribe to either of the extreme(-ist) ends of the swing of the pendulum: I think, therefore I am is arguably just as bad a place to be as the emotions-or-bust end, I feel, therefore I am.

The key is the word "incorporate" - (or include, engage with ... etc) - both ideas and emotions The Descartian 'rationalism'-or-bust modernism that bulldozed its way through the last half of the 19th Century, the two world slaughters and the cold war of the 20th Century, not to mention millennia of colonialism, has little to recommend it. And it's still alive and well in the 'cold' non-war of US, Russian and Chinese 'communist' neoliberalism, or monetization-or-bust, which reduces Descartian rationalist thought even further to naked numbers, with the London and New York stock exchanges both ending 2016 triumphantly (yes, 2016) on historical highs, and China making a serious bid to buy out the whole of the UK football premier league.

Neither of these extremes "I [only] think or feel, therefore I am" are worthy of the term 'abstraction', as both arrive at their core 'values' (sic) via a process of subtraction, philosophical reductionism, evolutionary regression, and paternalistic over-simplification.

The opposite to this hubristic reductionism (or physics-envy - i.e. the mistaken idea that everything is, ultimately, amenable to control - whether it's rationalist or emotionalist) is complexity, or complex-adaptive systems (think of it as dynamic ecologies). Its where most of us spend most of our time, living our lives forwards and making sense of them backwards. So one would think that the media commentariat would welcome it as the 'new' realpolitik.

Not a bit of it. John Harris writes that "complexity... is a 21st-century leitmotif, captured in those news-channel screens on which scrolling tickers and stock-market data combine to create the impression of a world so elaborate it is beyond anyone’s control. The average browse on Twitter creates exactly the same impression. ... Perhaps more crucially, individual lives are surely more scrambled and complicated than they have ever been. For a lot of us, in fact, modernity is a mess: not just of multiple user accounts, [etc] ... but working lives that now have to undergo endless peaks, troughs and reinventions."

And he titles his article "A society too complex for its people risks everything". But the point is not that society has all of a sudden become too complex for people (how paternalistic can you get?), but rather that the scales have been lifted from (some) people's eyes; people who for centuries have been assured by the political classes (via the media!) that life is a straightforward narrative; predictable, certain, simple, and stable - if you'll just follow me into the next global war/ recession / banking crisis (we do everything on a global scale nowadays!).

  • The shocking news is not that society has become too complex for its people, but rather that the underlying tendencies of biological and social life (from viruses and bacteria to multidrug resistance, onwards) have always been complex-adaptive. Life has always been more scrambled and complicated, and dynamic, and further beyond the reach (or conception) of any single big 'man' / big leader's /big media mogul's control than we have admitted.

  • Identity has always been a task, and a struggle, not a given (see Bauman, in section 2 & 3, here) for at least half, if not most of humanity. The issue is still coming out of the closet, as slowly and painfully as ever. The idea that gender, for instance, is a social [not biological] construct emerged many years ago with the first feminists, but widespread acknowledgement within and across many countries is still nowhere in sight.

  • However, we have always taken comfort from being told that life is simple and predictable - (although there are usually more than one versions of what this actually means) - and we seem to be falling deeper and deeper into this Alice-in-Wonderland rabbit hole. (Apres Trump, Marie le Pen?). Perhaps we have always lived in a fundamentally post-truth (or pre-truth?) society, wrapped in the swaddling clothes of either hyper-rationalism or hyper-tribalism.

Globalised-democracy, on the other hand, is currently in intensive care, in danger of being discarded as an 'end of history' contradiction in terms. (See the proposition quoted elsewhere in this project that "fascism is too exciting, communism is too boring, and democracy should be just right").
Simples .jpeg

All of this means, ironically, that the current crop of emotionalists are trying to drag us back into a world where all we have to do is "take back ('rational') control", and "make ... (fill in the blank) ... great again". As the meerkat says in the 'Just Compare' advert, its "simples". That's the rub. It's not, it's complex.

But as Trump knows only too well, the quickest, easiest and cheapest, simples-2-simples communication, is to reduce everything to tweets. He just takes the 'sound byte' to its logical conclusion. The whole media-tariat, which has lived high on the hog for years by promoting, and aiding and abetting the culture of sound bytes, must take responsibility for subconsciously daring someone to do just this (it sells like nothing else ever has) and push the ultimate sound byte over the edge. It might be Trump's nemesis, but that will just gift the media-tariat yet another story to sell.

Click-bait and echo chambers are the new memes and temes of simples-2-simples communication, or recursion masquerading as resonance, and even as abstraction. And twitter is in no way 'responsible' for it. Technological determinism went out of fashion sometime in the middle of the last century.

The point about (all) social media is that they are potentially very open systems, brimming with affordances, and begging for (negative) constraints, to make them into self-organising and self-regulating complex-adaptive systems, instead of 'global sewers'. If wikipedia could do exactly this, why not twitter, facebook etc. (To borrow a phrase: "The answer, my friend, is blowing in the spreadsheets of billionaires bank accounts" - wikipedia is not [yet] monetized). [WIP: add link to twitter issues ....]
[See also the section on Convenience Data, here].


1. Modal (i.e. sensory modalities: auditory, kinesthetic, visual, etc), to cross-modal (or patterns of patterns across sensory modes), to a-modal or modality-free (see Ramachandran's work and the examples in the Synaesthesia paper) (add links ... )

2. Interchangeable, token, or transactional signs: signs which have 'no' intrinsic meaning of their own. For example, 'I' and 'you' which are most commonly used as 'tokens' are 'passed' around conversations from one person to another, and signify different people, depending not on use per se, but on who is using that token at a particular time and place.

3. Signs which are, inherently, more abstract. Signs of number, for example.
Antekithera machine.jpg
Antekithera machine

This is perhaps (also) a subset of the next type.

Numbers are possibly also abstract in another sense, i.e. in the sense that the Athenians had
good reason to believe that prime numbers (which in a way stand apart from (are 'naturally' abstracted from?) the rest of the set of numbers - a notion which seemed to be supported by the stunning application of prime numbers in world's first computer - the //Antekithera// machine, which could predict eclipses (and many other things besides). (And then there are imaginary numbers, too ...).


4. Signs which are more complicated, as they become texts. These progress from gestures and words (roughly 'indexical' signs) to sentences, to paragraphs, to articles and books, etc, as their use moves from 'indication' to 'predication' for example the shift from an instruction to my dog to 'go outside' to a statement to my child that 'I would like you to go outside'.

Cross-contextual
As signs progress from indication to predication, to full blown texts, they gain the potential to be used across contexts - used in a different time, place, language, and era, although they do not always 'translate' that well. These sets-of-signs or texts are cross-contextual, in a similar way to the idea of cross-modal abstraction (see type 1). There is an additional layer to the analysis of this particular modality, which is Blackmore's gene / meme / teme analysis of our 'Pandoran' (and 'viral') society (add links). So in a variety of ways, signs, texts, and memes/temes take on a 'life of their own', independent of their original production, just as RNA / DNA / genes do in biology.

5. Context-free
Signs like 'time' and 'motion' are used (and thus have meaning) in everyday language, as well as in physics. In physics they have been highly formalised, resulting in much narrower meanings (and uses), which are stripped of subjectivity and of context. Here the subjectivity that is at the heart of 'predication', personal pronouns, and identity is set aside in the interests of a very different type of semiosis. This is deliberate, and methodological - not contingent or serendipitous (as is the case with cross-contextual signs and texts (see 4, above).

The result is a set of signs that is inherently abstract - its pragmatics is context-free (which is very paradoxical), as these signs and texts are constructed so that they can be used by anyone, anywhere, in any language, in any era - which makes them much more powerful, as well as potentially much more dangerous (e.g. the publication of the genome for anthrax). They are also, ironically, inherently falsifiable, which is the foundation of their (contingent) 'truth'. They can be seen as the set of meta-semioticsigns and texts (add links ...), a set which to various degrees share the essential characteristic of being context-free, i.e. the set comprising science, mathematics, bureaucracy, money, and representative democracy.

6. Complex-adaptive systems
Crucially, these systems are ordered yet unpredictable, although they do require (negative) constraints. (add detail ... )
(Note: this will probably all be deal with under section 0, above).

7. Opaque
These signs are too complicated to interrogate - they operate at a scale (size, number, scale, resolution, speed) which is useful - they can run complicated systems and perform very complicated actions - but it is impossible to interrogate their workings, as they work too fast for the human brain / interface. The four-colour map problem is a case in point, as is the case of current versions of machine 'language' interaction with humans.

And algorithms provide the next level of abstraction/obscuratanism - we could paraphrase Jensma's "I can dominate you if I know more words than you do" by updating it to "I can dominate you if I have more powerful (and pseudo-transparent) algorithms than you do" - see the cases of Facebook and Google [add links], and the influence of the Koch brothers [add link] - particularly in the Trump election campaign and (soon) presidency.

8. Viral / collaborative / cooperative
Within the overall description of knowledge as the capacity for effective action, and meaning as use (See Wittgenstein) the description of knowledge/s is somewhat dependent on structure (e.g. types of abstraction), and somewhat dependent on user-group, or discourse community (add links - Text and Discourse, Foucault, etc). Discourse communities can be highly informal (and random, even including avatars, nowadays) right through to the highly formal. The more informal they are, the more they can be viral, and based on emotional rather than cognitive criteria. On the other hand, uses too can be viral (see Blackmore's examples of memes and - in their more technically embedded form - temes (add link), and therefore more scientific.

9. Addictions
There are a number of addictions that would appear to be yet another category of abstractions. See here ...

The point about addictions is that they draw your attention and energy away from the messy materiality, and interconnectedness, and complexity, of daily life. Bertrand Russell, for example, wrote (in The New Quarterly, in 1907) that mathematics, "the world of pure reason, knows no compromise, no practical limitations, no barrier to the creative activity embodying in splendid edifices the passionate aspiration after the perfect form from which all great work springs" ... only to recant, in 1975 (in The Retreat from Pythagoras):

"One effect of [the First World War] was to make it impossible for me to go on living in a world of abstraction. I used to watch young men embarking in troop trains to be slaughtered on the Somme because generals were stupid. I felt an aching compassion for these young men, and found myself united to the actual world in a strange marriage of pain. All the high-flown thoughts that I had had about the abstract world of ideas seemed to me thin and rather trivial in view of the vast suffering that surrounded me. The non-human world remained as an occasional refuge, but not as a country in which to build one’s permanent habitation.


"I have no longer the feeling that intellect is superior to sense, and that only Plato’s world of ideas gives access to the ‘real’ world. I used to think of sense, and of thought which is built on sense, as a prison from which we can be freed by thought which is emancipated from sense. I now have no such feelings. I think of sense, and of thoughts built on sense, as windows, not as prison bars".

So we can probably describe addictions as i) a type of abstraction (type #9, provisionally), and/or ii) as mal-formed abstractions - perhaps for the simple reason that their 'use' eats itself up, and they are not sustainable (unlike other abstractions which provide - or aspire to provide - universal currency, and even universal recombinant currency). In the world of abstraction types, addictions deliver only malnutrition, no?

There is no easy way out of these paradoxes, or the schizophrenia of living in a world of mathematics which can be both an abstraction and an addiction at the same time. Semiotics does not provide any 'straight lines', but rather, endless loops (strings?).

And, as unsustainable forms - of malnutrition - addictions fail to achieve Monbiot's dictum that it is "by living lightly, we enrich our lives"

10. Derivatives
We are all just derivatives …
Biologically, socially, culturally, genetically …
But there are good, bad and ugly derivatives, and many feral ones too …

Convenience data is a form of derivative - at heart, just a metric (the commonest form of abstraction/substraction) - which differentiates things by counting and comparing just the amounts of things in various categories of difference- and their comparative ratios, rates of change, tendencies, etc (all the good things of mathematics and statistics). The pity is, in it's 'lite' or de-nuded form as convenience data, it's reductionist and simplistic.

Some potential case studies of extreme derivatives ...
  • The sub-prime mortgages of the 2008 crash
  • The Blair-Bush dodgy Iraq Dossier
  • Austerity as a one-line economic 'policy'.
  • Boris (Johnson)'s Brexit Battle Bus
  • "Outsourcing' as as a proxy for non-responsible management - (see the British Indian Empire [add links], what Robin Cook, the former UK foreign secretary and leader of the house, called George Bush's 'feral capitalism', and the ongoing privatisation of social services in the UK).

Some of these fall on the borderline between derivatives and dodgy metaphors, or just Goebbelsian lies e.g.
  • Donald (Trump)'s Mexican Wall
  • Israel's wall of 'self-defence'
  • Apartheid's homeland borders of 'self-reliance', or 'separate development'.
  • Pervitin.jpgHitler's denunciation of Jews as drug addicts, while he ordered 35million tablets of a derivative of crystal-meth - which they called Pervitin -(what an unfortunate pun it is in English!) - for the Wehrmacht's Blitzkrieg - or 'rave-krieg' one might say - see Norman Ohler's new book, Blitzed.


11. Silence // The 'Art'/efacts of Silent War?
The silence of drones strikes in the Nevada control room is an iconic (and who knows, paradigmatic) example of the new abstractions. (See: The Silence of the War, and Speaking of Agency). The drone strike is a crucial case of the 'de-articulation' of experience - a particularly modern trope of abstraction, which probably resonates with the downsides of 'out-sourcing' and with the tropes of 'objective' bureaucracy, etc. see ... [add link]).

And a new perversion of the Zen puzzle: 'what's the sound of one drone explosion?' In the control room, it's silence, and in the nuclear armageddon, its Rachel Carson's 'Silent Spring'. Some disturbing resonances there - if silence can resonate.