Contra Suspended Reason on Magic

Since the release of his “Fool’s Gold” essays (Pt 1 & 2), Suspended Reason and I have been trying to pin down a divergence between our worldviews/philosophies/orientations/methods.

Loosely, it’s a disagreement about how to talk about “magic” and “enchantment”. On a deeper level, though, it’s a disagreement about pragmatism and representationalism: whether, and to what extent, one ought to let words like “truth” and “reality” adhere to social or asocial aspects of the world. We’ve each tried to prosecute and precisify our arguments in various forms; however, I’ve been lazy. I haven’t generated much quotable content. Given the extent to which Suspended Reason is a textual animal, my oral-culture tendencies seem unfair. I retreat to conversation, providing only gestural and impressionistic sketches of my position; in so doing, I force him to rely on his own notes and recollections.

No more.

What follows is a first pass; a zeroth draft attempt to render my current sense of what we actually disagree about in prose, rather than generating yet-more rambling zoom calls.

Take the words that follow for what they are: notes, in search of the notable.

Agreements

Here’s some (expected) common ground.

Moderate Pragmatism

First, I think it’s useful to distinguish between two versions of a “communication remakes the world” claim, which I want to call, very roughly:

1. (Extreme) Metaphysical Pragmatism: There is no such thing as a non-intersubjective or non-subjective ‘reality’. An agent can cause arbitrary transformations of ‘reality’ via communication and belief, because talk of ‘The Real’ only ever indexes ‘things that the agent believes’ and/or ‘things that some set of agents believe’. In a very loose sense, this means that a given truth claim can only be evaluated with reference to its consistency with other claims ‘internal’ to a worldview, never w/r/t the ‘external’ validity of premises.
2. (Moderate) Ontological Pragmatism: ‘Knowing’ things about reality is fundamentally entangled with interaction with reality (including other agents in reality). Language and belief do not (and cannot) ‘represent’ the world in a perfectly neutral way; instead, they’re generated and evaluated in terms of their usefulness to action. ‘Objectivity’ consists in establishing ‘consistency’ of observation given layers of contingent/conditional abstractions.

I basically take it to be the case that we both agree about some version of (2), and our disagreement is centred on nuances of that view. Recently, I endorsed a version of (2) in my comparative discussion of state-of-being verbs. At a high level, I think that Suspended and I may be recapitulating a ‘Peirce v Dewey v James’ dispute about what exactly constitutes a ‘truth’ claim in light of a pragmatic orientation. I also think that we both think most versions of (1) are basically insane, and can be refuted with either (a) an incredulous stare, or (b) a gesture towards oncoming traffic and a suggestion that one simply ‘stops believing in the harmfulness of getting hit by a truck’ before stepping into it. In this sense, I think that even the most radical-seeming PoMo folks are twisting sticks while are ultimately endorsing versions of (2). I have no reason to think that Suspended endorses any version of (1).

Against Representationalism

Second, I think we probably both agree with a stylised critique of “representationalism” in various specific domains. While I’m not enough of a Rorty (or Putnam, or Brandom, etc) scholar to give an unqualified definition this theoretical assumption, I’m comfortable with sketching a version of ‘representationalism’ (and its flaws) for a specific domain: the use of models in science. Here, the basic representationalist premise is shaped as follows:

1. Models either
   1. themselves ‘represent’ (features of) an already-given external world/reality, or
   2. enable us to ‘represent’ (features of) an already-given external world/reality.
2. The ‘representing’ is an inherent quality of either
   1. the model, or
   2. the modelling process.

On the representationalist view, broadly speaking, there are two conditions which some objects satisfy such that they are models:

1. An Aboutness Relation: the model objects stand-in for, and are in some relevant sense ‘about’, some other real objects in the already-given real world.
   • Most often, uncertainty about the details of this relation is seen in a single question: if a model represents its target, in virtue of what relation does this representing occur?
     • Some people call this the “enigma of representation” (Frigg 50).
     • Other people call this “the crux of the problem of representation” (Morrison 70).
     • FWIW, Suárez sets a pretty clear limit on our need to clarify this relation. For the purposes of explaining conditions for an aboutness relation, the goal is to answer “In virtue of what does a certain model represent a certain system?” not the more general question “In virtue of what does a certain model represent a certain system accurately or truthfully?” (767–768).
   • The usual explanation of the relation is via some denotative relation (ie, a dyadic relationship between model object x and modelled object y). For example:
     • R.I.G Hughes uses a ‘Denotation, Demonstration, and Interpretation’ framework to claim that denotation establishes a representing relationship between model and reality (see, eg, here and here).
     • Some structuralist accounts construe a denoting relation for surrogative reasoning (see, eg, Contessa).
     • Frigg & Nguyen advance a ‘Denotation, Exemplification, Keying-up, and Imputation’ framework which also essentialises denotation.
2. A Comparable Epistemes Property: model objects are representations which are studied for the purpose of learning about their targets; they facilitate the acquisition of knowledge.
   • This assumes that two cleanly identifiable steps are possible (and actual):
     1. We study a model system to elicit new information about its properties and relations (ie, we generate claims about the model system).
     2. We convert our new knowledge of the model system (from the previous step) to knowledge of the target system, by way of comparative claims. These comparative claims are based on similarity (or other) relations. That is, we rely on a claim that the model ‘shares some properties’ in ‘some respects’ or ‘to some degree’ with the target, and so transfer model-claims to target-claims.
   • Whatever this property consists in, it’s taken to be a condition which object x must satisfy in order to be an epistemic representation of a distinct object y. The model stands in a representing relation, and it’s in virtue of this relation that knowledge of the model that enables knowledge of reality.

So sketched, the ‘representationalist’ view has a lot of stones in its proverbial pockets: problems and vague definitions, most of which appear to be persistent in the face of inquiry. Collectively, these are, on my view, likely fatal. For example, the ‘aboutness relation’ assumes denotation, but denotation is itself suspect: in order to say that an object x stands in a denoting relationship to another object y, you need to stably identify both the existing objects — the denoting symbol and the denoted target — and their properties; however, object and property identification is usually vague to at least some degree in practice. You also need to explain why the relation isn’t symmetrical (why doesn’t y denote x?) unless you’re comfortable biting a lot of weird bullets. Likewise, the ‘comparable epistemes’ property fails to specify bounds on comparability. Presumably one wants something more permissive than “x is comparable to y iff it shares all and every property with y” and less permissive than “any thing is comparable to any other thing”. A failure-to-specify kicks the epistemic can down the road, with catastrophic results: representationalism so posed isn’t much a coherent description of modelling as it is a reformulation of what needs to be described.

Underpinning all of this, though, is a more fundamental counterargument, which I expect that Suspended and I both endorse: talk of ‘objects’ representing ‘objects’ — or of ‘language’ representing ‘things’ — buys into what one might loosely call “the myth of the given”. It is at root a kind of Cartesian Dualism. It assumes (but fails to specify reasons to believe in) the possibility of unmediated, unbiased, impartial, omniscient, and separate observation of ‘objects’ which always already exist; objects which may need labelling, but which nevertheless have definite boundaries for which there are pre-existing, ultimate, and universal facts-of-the-matter waiting and able to be discovered. This is implausible. Such stability seems unlikely to be true for anything but ontological simples. It’s also not how scientific “observation of reality” or “prediction of reality on the basis of abstractions over observations” is generally taken to function in practice. In order for me to believe that ‘ordinary’ objects ‘exist’ in a deep sense — that is, that there is a definite thing as a ‘chair’ or even ‘this particular chair’ which precedes an agent’s entangled transactions — I need to believe a lot of things that are inconsistent with materialism. Carvings-up-of-reality at any level of granularity higher than simples are in some sense arbitrary.

Minimal Ontological Commitments

A more coherent account how we talk about objects and their reality — and how we talk about model abstractions and their accuracy — is to strip the discourse of its (apparently unqualified) ‘deep and fundamental’ truth claims in favour of a more minimal view. I expect such a view sound something like:

When people say that some claim is true, or that some object exists, they almost always mean that it’s conditionally true given a massive corpus of unstated premises, including such-and-such a set of fuzzy anticipations about experiences/sensations, and such-and-such a set of could-be-otherwise judgments which index connections between or generalisations across common features of experiences/sensations.

Such an account breaks fewer intuitions. It tracks with a number of aspects of experience, cognition, and communication to which I (and most humans) have privileged first-person access. It suggests that many persistent disagreements in ‘metaphysics’ are empty, in the sense that they are confusions caused by messy and imperfect language. And it also gestures to the way that free-floating could-not-be-otherwise truth claims made in a priori abstract deductive logic are not the same as truth claims made about trees or cars — or Eli Hirsh’s ‘incars’ and ‘outcars’ (32) — or atoms.

I take the minimal account that I’m gesturing towards, here, to be something like something that some people would like to call “pragmatism”. I think that Suspended and I both endorse it.

Disagreements

Where I think we diverge in our thinking is how we each think that one should — when communicating precisely and intending to be understood! — handle talk of ‘truth’. On one side, there is the suggestion that one should reserve (unqualified) talk of ‘reality’ for the deep and definitive could-not-be-otherwise claims: the non-subjective, non-intersubjective, reality-bites-back stuff; and, separately, the a priori deductive logic stuff. On the other side, there’s the suggestion that should instead basically just bite the French PoMo bullet.

My issue with some talk about the “enchantment” of the world — some talk about the way that language and communication “remakes the world” — is twofold: first, it mistakes a methodological claim for a metaphysical one; second, it’s bad strategy.

Methodology, not Metaphysics

Recognising the contingency of human ontologies is an obvious first step to understanding and ‘explaining’ culture. This is true when you’re interested in answering the comparative questions — “why are these representations more successful than others in a given human population?” (Sperber 49) — and also when you’re asking the famous interpretive ones:

Believing, with Max Weber, that man is an animal suspended in webs of significance he himself has spun, I take culture to be those webs, and the analysis of it to be therefore not an experimental science in search of law but an interpretive one in search of meaning. It is explication I am after, construing social expressions on their surface enigmatical. (Geertz 5)

Groups of humans share ontologies in order to coordinate. Moreover, whatever ontology a given culture settles on, that ontology is necessarily going to be somewhat value-laden, and somewhat historically (and environmentally) arbitrary. It will also be a stack: at a high level, a given culture’s ontology will necessarily diverge with lower-level ontological factorings. To gain in generality is to lose in faithfulness. Even the lowest-level of human ontologies are pretty general, and not particularly faithful to The Real.

But investigating contingent ontologies is an ethnographic and sociological project, not a physics. As Holbraad and Pedersen remind us in their preface to The Ontological Turn (2017), “the ontological turn in anthropology must be understood as a strictly methodological proposal – that is, a technology of ethnographic description” (ix). Such a project “is not concerned with what the ‘really real’ nature of the world is or similar orthodox philosophical or metaphysical agendas often associated with the word ‘ontology’ (x). Rather, it is a project in which,

operating always as an adjective or adverb — never as a noun! — ‘the ontological’ here is meant as a call to keep open the question of what phenomena might comprise a given ethnographic field and how anthropological concepts have to be modulated or transformed the better analytically to articulate them. To take the ontological turn is to ask ontological questions without taking ontology as an answer. It is in this sense that it represents an intensified, more thoroughgoing commitment to more traditional forms of anthropological reflexivity, rather than a rupture with them. Instead of closing off the horizon of reflexivity in the name of some sort of ultimate reality that may ground it (an ‘ontology’ in the substantive sense), the ontological turn is the methodological injunction to keep this horizon perpetually open. To recall Clifford Geertz’s (1973: 28–9) invocation of the old adage about turtles (though he of course was writing about culture), this intensified manner of reflexivity goes all the way down. (11)

As Eduardo Vivieros de Castro put it in “Who is Afraid of the Ontological Wolf?”,

in an important sense, what we have been advocating was essentially an ontological turn of the epistemological screw – a methodological tightening up of our ethnographic descriptions, which, rather than allowing us to ‘discover new things’ about the other, marked the limits – ontological, not critical – of what can be known (and then said) about that other. (10)

To what extent does Suspended’s project draw always and only on ‘ontological’ talk as “a technology of ethnographic description”? To what degree is ‘the ontological’ operating, for him, “always as an adjective or adverb” and “never as a noun”?

In day-to-day life — which is to say, when he is buying food for his cat or beer for his friends — I do not expect that he is actually confused. He’s a smart guy! He functions in the world!

My issue lies only in his admittedly vibrant prose. In “Fool’s Gold, Part 1”, Suspended writes:

But there is also a California Metaphysics, not so much a coherent position as an axis of conflict between NorCal and SoCal, its Mason-Dixon cutting somewhere between Pismo Beach and Los Olivos. Silicon Valley4 and Haight-Ashbury versus Hollywood and the Elysian Heights. Santa Cruz joint smokers clad in drug rugs, versus the bronzing dab-men of Venice Beach. In the balmy Mediterranean climes south of Pismo, a classical attitude towards masks and persona reigns. Personhood itself is a front, a face, a performance. Where a NorCal hippy might grow out body hair in a back-to-nature move, some reclamation of lost authentic states, the SoCal hustler sees one more fashion trend, one more set of symbolic postures. (“The Jesus-look is in.”) Peel away the surface, some Californian metaphysicians (shrugging) tell us, and you’ll just find another surface. The landscape of the interior is one more social strategy, a disposition in service of performance. Or—if some hidden and privileged interiority is ceded—it’s looked down upon the way a nineteenth century cranium-measuring aristocrat might condescend to natural impulse. Surfaces are ennobling, aspirational. Get back to nature and all you’ll find are animals.

Not that either camp are proper relativists. The promise of psychedelics lies in their ability to de-naturalize the ready-at-hand, to present alternate surfaces and make explicit our structuring interpretive schemas. Beach-bum SoCalites are liable to hedonistic languor, soaking in the brightened colors and tracer viz, tripping just one more performance. NorCal hippies, though, are liable to take those new surfaces and slogans as deeper and underlying truths, to take every new perspective as “the” perspective. Until you dose, you just won’t get it. LSD is paradisal, the trip either life-changing inflection point, or the sort of temporary glimpse of Eden’ll drive a man insane just chasing it forever. Everything changes—until the next morning, when the newly converted fall back on old habits or get reabsorbed by intelligent social webs. Because men are made by their moments, habits are solutions to problems, and none of those problems have gone anywhere.

Beneath the surface, another surface; beneath the mask, another mask. This is ‘true’ for the French aristocrat when times are good. We are all of us much more aristocratic than we were a century ago. But Pynchon, for all his enchantment, and all his tragicomic genius, is almost always describing the mindset of the paranoid. His work is often about the interior life of people being traumatised and transformed into better postmodern subjects. Gravity’s Rainbow is a classic for precisely this reason. The “Proverbs for Paranoids” are transparent:

You may never get to touch the Master, but you can tickle his creatures. (237)

The innocence of the creatures is in inverse proportion to the immorality of the Master. (241)

If they can get you asking the wrong questions, they don’t have to worry about answers. (251)

You hide, they seek. (262)

Paranoids are not paranoids because they’re paranoid, but because they keep putting themselves, fucking idiots, deliberately into paranoid situations. (292)

Compare these to fragments from Suspended Reason’s prose:

Delusions make themselves true, or carry metaphorical truth, or are necessary means for bringing communities together.

There is no true world, nor a true lodge. No paradise, no panacea—only more flawed worlds.

If I take Suspended’s project to be a literary-poetic one, I have no disagreement. But the apparent slippages concern and confuse me. Are these claims about what aristocratic lifeworlds feel like from the inside? Or are they claims about mechanisms which (on his view) necessarily underpin all modes of human meaning-making? The former is ethnographic; the latter is suspect on my view.

And this brings me to my second disagreement:

Strategy (and Ethics)

In “The Engineer and The Diplomat”, Ben Hoffman suggests a compelling intuition-pump. In full:

Imagine you’re an engineer, at an engineering conference. You see a group of your colleagues, including someone working on your project, in an intense conversation. They’re asking your friend questions, and your friend obviously has to think hard to answer them. Your intuition is going to be that something good is happening here, and you don’t want to interrupt it.

Now, imagine you’re a diplomat, at a diplomatic conference. You see a group of diplomats, including someone representing one of your allies, in an intense conversation. They’re asking the allied diplomat questions, and your ally obviously has to think hard to answer them. Your intuition is going to be that something bad is happening here, and you want to derail it at all costs.

I think that most people perceive most conversations as more like the diplomatic conference, than like the engineering conference. Challenging someone’s narrative is not a way to improve everyone’s world-model and make everyone better off; it’s a social attack against the person being challenged, a bid to lower their status and exclude them from the group. And people who want the group to get along will try and rescue the person being challenged. To divert attention from the stress-point.

This is mistake theory and conflict theory in sociological microcosm. It’s also the problem I have with Suspended Reason’s enchanting accounts of enchantment. He sides with the diplomat. He shouldn’t. The diplomat is doing harm.

Somewhere along the way — perhaps during the world wars — the institutions of industrialised Western Culture pivoted. Over period of around a decade, “the engineer’s concept of feedback was introduced into the social sciences,” and two tightly-entangled lines of work developed (Richardson 92). One thread generalised the theories of servomechanisms to the economics of production control and industrial dynamics, and noted the implicit ‘feedback’ thinking of existing economic theories. The other thread placed greater emphasis on ideas of self-reference and homeostasis. Early work was exciting enough to prompt the famous Macy Conferences, which ran from 1946 to 1953; today, we know the unified field of thinking that resulted as cybernetics.

For all its value as a framework, however, “cybernetics” was from the outset deeply entangled with a project to generalise the insights of industrialised and mechanised warfare. Kurt Lewin, inspired by the “multitude of self-steering missiles” that had been developed during the Second World War, sought explicitly to “understand what kind of social organization is required for efficient planned group action” (442). He wanted “research for social management or social engineering” and a project of “action-research” aimed at outcomes which transformed society (443). Norbert Weiner’s Cybernetics (1948) sketched the observation

that the social system is an organization like the individual, that it is bound together by a system of communication, and that it has a dynamics in which circular processes of a feedback nature play an important part. (24)

And by the time he published his non-technical follow-up, The Human Use of Human Beings: Cybernetics and Society (1950), the program was pretty clear in the title. Karl Deutsch’s The Nerves of Government (1963) is similarly transparent. Cybernetics was about information-action feedback in systems composed of humans.

My criticism, here, is not intended to be genetic or geneological per se. The fact that “cybernetics” thinking developed out of the mass-mobilisation and industrialisation of World War 2 does not disqualify it. Rather, what I want to highlight is the fact that cybernetics brought with it a novel view of human agency and subjectivity.

Nowhere is this clearer, I think, than in the work of Talcott Parsons on what he called the “cybernetic hierarchy”. In reflecting on the development of his own thinking, he wrote that

Clarification of the problem of control, however, was immensely promoted by the emergence, at a most strategic time for me, of a new development in general science—namely, cybernetics in its close relation to information theory. It could now be plausibly argued that the basic form of control in [human] action systems was of the cybernetic type and not primarily, as had been generally argued, the analogy of the coercive-compulsive aspects of the processes in which political power was involved. Furthermore, it could be argued that functions in systems of action were not necessarily “born free and equal,” but had, along with the structures and processes implementing functional needs of the system, differential hierarchical relations on the axis of control. (850)

The question was and is essentially one of control: what modes of social organisation would tend to win out. As Ronald Kline puts it in The Cybernetics Moment (2017),

Parsons applied the cybernetic insight that systems high in information but low in energy (like the household thermostat) could control systems low in information but high in energy (like the household furnace). (148)

A social system structured in the ‘cybernetic mode’ relies on an individual human subject which is, as much as possible, predictable in a way conducive to smooth functioning within a broader mechanism. There is a cybernetic apparatus in operation; what Foucault might have called a dispositif (cf. Deleuze). It is

a strategic system of relations established among a heterogeneous ensemble. Institutions, architecture, scientific and moral statements, and instruments were among elements that, in response to an urgent need, might be organized into an apparatus. (Geohagen 99)

When Suspended Reason gestures enchantingly to a grounding of “truth” and “reality” in the social, he sides with the diplomat and against the engineer. He writes from a cybernetic frame. This is fine, I suppose, but it carries with it a fraught notion of the individual as a moral subject and agent. And for all its present dominance, this cybernetic apparatus is not — need not — be universal.

“Culture is a shroud,” says Robert Watson, writing about Hamlet (214). In one way or another, every culture has a “fabric of denial (however richly brocaded) beneath which it hides its dark obvious secrets.” (214) Marilyn Strathern might have put it another way: no nature, no culture. Culture is that which we take to be natural.

In his treatment of magic, Suspended takes the most suspect version of a “cybernetic” dispositif to be natural. On current margins and ethical grounds, I think he should resist such a culture.

Delusions are delusions; they are not a “necessary means for bringing communities together”. Descriptively: only some communities; only sometimes. Prescriptively: only the harmful ones.

(See? I, too, can twist the stick.)

Dissolving The Disagreement

In one of our conversations, Suspended gestured to a delightful hypothetical: fencing in a cloud of fog. If I trust my shoddy teeline shorthand, the setup was as follows:

If you’re in a cloud of fog, and I stab you, you’re going to bleed and die. This is a biological fact, and it doesn’t matter if nobody saw me stab you. However, if we’re fencing, and there are judges, it’s extrinsic, in Goffman’s parlance, insofar as the fog generates a confusion. If we’re fencing, the question of ‘the consequences of me stabbing you’ centres on the engagement of human perception, and how important human perception and belief are to outcomes. The squishiness of social construction is the relevant thing, because changing an agent’s beliefs changes the world.

I suspect that this hypothetical reveals something deep about our apparent disagreement: it’s all mere terminology.

If the two of us were face-to-face, épées in hand, I don’t think that Suspended thinks that the presence of a fog machine would prevent him from killing me. I don’t think he thinks there would cease to be a brute ‘fact of the matter’ about whether a point was technically-if-not-in-practice scored. What seems more likely, to me at least, is that he is just more aristocratic than I am. He spends less time in contexts of harsh environmental constraint. He sees less of the intrinsic and more of the extrinsic about him. Nothing more.

As he himself puts it,

The form of language serves the social problems that language solves. Language is as language does; it’s a cultural practice, not an object; it’s a social technology; it can only act through its effects on agents. I can’t move a rock or build a wall with language on its own.

If more of Suspended’s environment is constrained by agent-manipulation — and less of it is rock-moving — then our disagreement over terminology makes sense. Our worldviews diverge in the same way as any other two people translating across neighbouring cultures. Our language is different, sure; yet for any and every concrete case, our anticipations are the same.