I've decided to start putting some of my blog posts from Philosophy Bear here as well, particularly those with special relevance to this community.

A curious fact about the English language

In the English language, cynical means both seeing the imperfections of the world, and being ruthless (“he acted cynically” said of someone who accepted a bribe). Innocent encodes the opposite dyad.

I have no idea how far beyond English this pattern goes. I would guess it is a deep tendency given that the tree that putatively caused the fall was that of the knowledge of good and evil. A brief search suggests that the pattern probably is common in Indo-European languages and appears to crop up in Chinese.

The underlying theory, presumably, is that when you see the nonsense everyone else is up to, how predatory the soul is, you feel free to break norms yourself. Another facet may be the simple fact that when you have more options for effective mischief and getting away with it, you’ll probably get up to more mischief. Finally, for some kinds of awareness, a greater capacity for rationalisation may be present, as well as more charitably a greater capacity for more sophisticated ethical views which might appear like licence to those of greater simplicity.

There is little empirical evidence that people of greater awareness, either generally, or of the faults of others, are more likely to behave anti-socially, even when doing so is to their advantage. There is evidence that a general belief that people are bad is negatively correlated with cognitive ability and bad behaviour, but this is not the same thing. We might usefully distinguish between epistemically justified cynicism - being cynical when the evidence calls for it - and general moderate edgelordism that is not epistemically justified.

There’s a concept in the psychological literature: Machiavellianism. If one looks at the question inventories, it runs together a bunch of conceptually distinct things including A) Cynicism, in the sense of a low baseline view of humanity B) Willingness to manipulate C) Awareness of how to manipulate D) Willingness to harm others in order to get what you want. These are all distinct conceptually1, and the task of exploring their conceptual and causal connections is often neglected. A special difficulty here is that there are reasons to think self-report will be problematic. For example, thinking you are good at knowing how to manipulate does not guarantee actual skill, and it’s easy to see how those who are okay with, or even proud of, their ability to manipulate might tend to overrate their skill in that area.

There is a concept a little bit like cynicism, but different, I am interested in. Call it instrumentality. Instrumentality is the constant awareness that different things you might say or do could be useful to make others act in certain ways and achieve outcomes. Instrumentality as a form of perception is frequently alienating but it is not, I think, evil. Further, anecdotal experience suggests to me it is extremely common in a certain kind of autism. It begins to develop in childhood, where it takes on a largely compensatory character. A usually automatic process of social calculation is run explicitly. I am interested in what it looks like in adulthood, where it is largely understudied, but where I suspect it often becomes somewhat useful - not just deficit filling. Of course, this is not to say that the overall picture isn’t one of deficits. Whether the deficits in this area outweigh the gains likely varies from person to person. A necessary disclaimer here- instrumentality is far from universal in autism and plenty of people who aren’t autistic exhibit instrumentality. Like almost all human traits it is a matter of degree, not kind.

I am interested in philosophising the phenomenology, ethics and psychology of instrumentality, and that will be our task today.

Since I was a child, I have had a constant, ticking part of my brain suggesting strategies to manoeuvre people around me to get what I want. I do not think that this means I am good at it- the evidence would tend to suggest I am somewhere between the 20th and 80th percentile. But it is always there. It makes me feel disengaged and inauthentic in a certain way. It is probably not a coincidence that I have moralistic obsessive-compulsive disorder. Although my OCD has rarely if ever focused directly on this awareness- most of the time I am relatively shameless about it, if somewhat sad that I am “cut out of the loop” of direct, “authentic” action.

Problems with traditional definitions

My view is that part of the moral squeamishness that exists around instrumentality arises from the messiness of our concept of manipulation. It was made for and by people who don’t see the world like this.

By a normatively parasitic concept I mean a concept that appears to have independent use for moral analysis, but is, at bottom, another judgment whose independent content beyond ‘good’ and ‘bad’ is limited. It is a normatively thick concept where the descriptive part is all too thin to pull analytic weight. I suspect the concept of manipulation is normatively parasitic because it’s too useful to play straight. We want it to be vague for our own benefit.

Consider, one’s ex can be manipulative for honestly stating what they saw as your good attributes - leading you on by being so liberal with truthful compliments. Letting you inaccurately form the impression that he was really into you. On the other hand, you are not manipulative for telling your boss the “white lie” that the traffic was bad when you slept in. Somewhere else, a man is grumbling that he was called manipulative merely for kind and truthful comments, and a boss is grumbling that employees these days think nothing of baldfaced lies after not even bothering to set an alarm.

It seems to me that manipulation is a threshold concept where the threshold in question is moral. Attempts to ground it on something like degree of epistemic distortion or bypassing reason hit the problem that the lie to the boss is totally distorting, but many (?most) would baulk from calling it manipulation. On the other hand, we have forms of manipulation that aren’t epistemically distorting, e.g. Noggle’s example of peer pressure. What about Noggle’s own account? Noggle’s attempt to ground the concept in leading astray from the persuader’s own values seems wrong for two reasons. Firstly, suppose you are really, truly a patriot and so am I. However I put enormous pressure on you to go to war for my own selfish benefit, that seems to me manipulative- similar to Noggle’s own example of peer pressure. Secondly a demon who believes in evil, and drips honeyed words to lead a mortal into sin is still a manipulator.

The best explanation of usage here is moral. Manipulation is vaguely manipulative stuff (lying, pressure, flowery rhetoric) which crosses a moral threshold, both in its inherent content (e.g. nasty pressure and mendacious lies) and in what it aims to affect2. The combination of unclear threshold and continual disagreement about what counts as pushing something towards the threshold, is what makes the concept normatively parasitic.

Quick aside. Above I invoked the concept of “vaguely manipulative stuff”. Using a word in your definition of it is, to say the least a warning sign. Thus I owe you an independent definition of “vaguely manipulative stuff”. I think what makes something a candidate for being manipulative if it passes the moral threshold is no one thing, but a family of often related things that philosophers definitions tap into- deceptiveness, bypassing reasoning, bypassing the agent’s own values, bypassing the persauder’s values, and also folk notions like “being too strategic”. They are organised by something more like a social role than an ethical category. These are then subject to a moral threshold to determine whether they are manipulative, and this is the best sense I can make of the concept. The concept might be socially rich, but it is, as I said above, thin on content that is useful for the moral theorist, and thick on direct judgement.

If manipulation is a moral threshold concept this is devastating for anyone who wants to make “was it manipulative” a direct criterion of rightness and wrongness. You can’t decide whether it’s manipulative until you decide how wrong it was, and you can’t decide how wrong it was until you decide how manipulative.

Another way we could go owes to Shlomo Cohen who argues that manipulation isn’t inherently bad- it’s often just social lubricant. Now we can simply adopt maximalism. White lies to your boss? Manipulation. Our complimentary boyfriend? Manipulative. Every boyfriend, girlfriend and theyfriend since the dawn of time? Manipulative. Me thinking carefully about how to put my mum in a good mood because she’s been having a hard week? Manipulative. This is also a legitimate way to go- we even have a built-in explanation for why we don’t normally use the phrase, conversational implicature. But once again, manipulation is dethroned as a useful principle for moral philosophy, or at least as a right-wrong yardstick.

I think a thought experiment is interesting here, one that shows our concept of manipulation becomes very odd very quickly. Josh is a pre-cognitive telepath. He always knows exactly what to say in order to get what he wants from people- if any such string of words exists, he will find it. But his situation goes deeper than that. He knows what every string of words he will say results in. He knows if people will say yes or no in all cases. Can Josh avoid being manipulative, however badly he wants to? On most of the philosophical accounts we’ve explored, probably, but these accounts as we have noted have their own problems. There is however a kind of folk concept of manipulativeness on which Josh is manipulative. That folk concept ties manipulativeness to thinking too strategically, essentially identifying it with instrumentality. I think Josh shows the moral bankruptcy of that concept, at least if “manipulation” is intended as an ethical judgement. If Josh is not always manipulative a certain kind of awareness or planning is not what separates us from the manipulator. For the reasons I have indicated, I suspect all that leaves us with is a normatively parasitic concept.

The person who sees the world instrumentally is just a lower fidelity version of Josh. But we are all, I think, just lower fidelity versions of Josh. We are all calculating, with varying degrees of success and conscious awareness how to produce desired outputs with available inputs, whether those outputs are wanted selfishly or altruistically. The main difference between the instrumental and the non-instrumental person is the conscious salience and availability of calculation not the presence or absence.

[The concerns I have raised about the concept of manipulation are, I think, transferrable to the perennial philosophical concept of “treating people as means rather than ends” How would Josh even begin to not treat people as means? Other narrower concepts like “objectification” might also run into trouble. This would require a much longer treatment than I propose to engage in here, but the concept has always been, to say the least, difficult, and Josh I think will make it more so.]

Earlier when I introduced instrumentality, I emphasised that it is cut off from willingness to manipulate. If we take manipulation as meaning something like wrongdoing through influence this is true, but if manipulation is a moral threshold category without a clear separation, the truth is not so clean. The instrumental person perceives a variety of strategic options and like most of us hopes they do not exceed the limits of the good. However, they are continually aware of the possibility that they are manipulating, that they have the moral calculus wrong, in a way that a less instrumental person might not be. From the inside this can feel terrifyingly like willingness to manipulate, if one is not careful.

The moral status of instrumentality

Now, our final question, is instrumentality morally bad, or can it be equated with ruthlessness? I don’t think so, and having spelt out what we have, I think the conclusion becomes almost unavoidable.

I do not think, to be clear, that instrumentality is necessarily a healthy way to live. I think it is, in a sense, a bit like constantly thinking William’s3 One-Thought-Too-Many:

[This construction] provides the agent with one thought too many: it might have been hoped by some (for instance, by his wife) that his motivating thought, fully spelled out, would be the thought that it was his wife, not that it was his wife and that in situations of this kind it is permissible to save one’s wife.

I am not denying, by any means, that my running calculations on exactly what to say to my weeping friend are costless. It is analogous to one-thought-to-many exactly, in that it is a kind of deliberation that threatens to erode the very same value it aims to preserve. It is a failure to embody something beautiful, a natural love unmediated by moral or strategic categories. Not a complete failure, not an all-the-time failure, but a frequent and recurring one.

However, it is unfair to treat instrumentality as paradigmatic of ruthlessness for two reasons, both of which are decisive: A) because it is unavoidable and therefore blameless by ought-implies-can4 B) because there is no evidence of increased selfishness, or behaviourally expressed ruthlessness here. Whatever defects of connection or experience arise from instrumentality might be tragic, especially at certain times, but that does not make the condition subject to condemnation.

I should say in completeness that there are benefits as well. Seeing the world in terms of levers of action allows a certain heroic stance. I do not want to praise myself or those like me, but if one is aiming at the good, there are benefits, both for the world and for oneself, in the highly agentic view that instrumentality provides. I have previously rejected Williams one-thought-too-many argument against consequentialism precisely because I think he misses that there is something oddly affirming of the beauty of humanity in being able to step away from the personal, to unravel the value of the specific on behalf of a larger principle. This is a topic I hope to revisit because I don’t think I did it justice last time. A preview: one might just as easily say that refusal to consider whether morally one should save one’s wife itself undermines the value such a refusal aims to preserve- makes one’s love an animal affection rather than expression of a human principle. Call it Platonism or Christianity but I think we are best when we are logos made flesh, and our loves are principles given particular meanings.

I also think there is a certain self-congratulation here around the non-instrumental view that is quite ugly when spelt out. I have managed to submerge my social games largely below the waterline of consciousness, you have not, I am morally purer. That is worth rejecting. Are we really free of one-thought-too-many if we have concealed that thought from ourselves?

Finally, and this is not my thought, but comes from a friend, perhaps there is more philosophical work to be done on the idea that in some circumstances conscious knowing is always alienating. We might refer to this as epistemic alienation. There’s a debate, I believe, in medieval theology about the disturbing possibility that God cannot fully appreciate music because he sees it from outside time. Perhaps analogous problems apply to humans all the time - knowing is already losing half the battle, at least if that knowing is in your attention, because it is constitutively incompatible with, or in deep tension with, certain kinds of experiences and perhaps even relationships.

Originally posted here on my Substack

Footnotes:

1

B & D might be conceptually identical on certain very restrictive definitions of manipulation, but most people, I think, do not find these definitions plausible.

2

Since I am an assessor subjectivist, this will amount to “The putative manipulator does some proto manipulative stuff, lying, rhetoric, pressure or whatever and from the point of view of the speaker-assessor this crosses a moral threshold”

3

I might have chosen as alternate interlocutors here Iris Murdoch or Martin Buber. Sketch is left as an exercise to the reader.

4

I have mixed feelings regarding whether ought-implies-can, but in the absence of a strong counterexample here I am inclined to think it does.