I recently listened to a debate, of sorts, to further develop the philosophical underpinnings of the concept of exploitation. The contenders were Hillel Steiner and Nicholas Vrousalis. While I believe the enterprise is important, I didn't like either of the formulations.


I found Steiner's formulation too convoluted. He imagines all exchange of resources as reducible to a kind of auction, which I'm sympathetic to, but then says that it's some previous injustice or infringement of rights on the part of either the seller or the buyer that prevents the resource in question from fetching a fair price.

That is, a concept of rights and the infringement thereof has to exist in order for exploitation to occur. I don't buy this because it's easy to imagine an exploitative situation where rights don't factor in. Indeed, Vrousalis replies with a scenario where one person has fallen into a pit, and another agrees to rescue him only on the condition that he pay some extortive sum or sign a bond of indenture. Steiner concedes that this may indeed be extortion, but only counts as exploitation if there is some preceding right granted to people who fall into pits that they ought to be rescued. You're welcome to listen to him explain his position further, but I couldn't understand why the distinction is significant.

This is part of the reason why I don't like the concept of rights to explain how people ought to behave to one another, and what people are and are not allowed to do. The concept is arguably a throwback to thinkers like Hobbes and an ultimately unsatisfactory means of expressing what's going on when the actors in question are theoretically supposed to be equal and want the same things. Rights are entitlements in a contract, and contracts are asymmetrical: Have A, want B. When there is only A to have or want, and all the actors in the system are sovereign, the agreement is called a treaty or protocol. But this is probably a subject worthy of its own treatment.


Vrousalis's position, on the other hand, I found to be tautological. He said that exploitation has to do with vulnerability. The concern that I have is this represents a narrow view of the concept of vulnerability, and conflates it with general precariousness. He uses variations of the general example of the pit to show that people who get into vulnerable situations are amenable to being exploited by others. Of course it does: you can't exploit the invulnerable.

We use the phrase exploit a vulnerability all the time in the business of information security. In that context, we're typically talking about piercing through a specific chink in a target's armour to gain control of a system or otherwise benefit with at best no regard to either benefit or harm to the counterparty. This is significant because it decouples the concept of exploitation from the sociopolitical/economic standing of those involved. Rather, it considers the context: is one party effectively taking a share of benefits that in a fair and just system would belong to another? Are they stealing?

Will as a Prerequisite

Also present was the issue of whether or not the beneficiary of a potentially exploitative situation had to be aware of their role. My feeling is no: it's easy enough to conceive of a person born into dynastic wealth, for whom free run of the planet, influence on public policy, and being waited on hand and foot is normal. Arguably, the OECD exploits the rest of the world under the regime of globalism, when only a tiny sliver of the billion-plus people who make it up might reflect that such a situation could even be possible.

The Interesting Position

The most interesting position came from the questions: somebody simply asked about the role of information asymmetry on exploitation. Neither Steiner nor Vrousalis had anything particularly useful to say in reply. In fact, their responses implied neither really understood the question, or even what information asymmetry means.

This is disappointing, because I think information asymmetry is a much cleaner candidate for characterizing exploitation than either of the two initial formulations. Information asymmetry is simply one agent in a system possessing a fact about the current state of affairs of material consequence that another does not. It accounts for Steiner's auction scenario as well as Vrousalis's pit, and does a better job of both.

For Steiner's auction, we can get rid of the idea of a prior infringement of rights, and replace it with information asymmetry: Say the buyer knows something about the offering that the seller doesn't, or the seller knows something the buyer doesn't, or there is a potential higher bidder out there that doesn't know the auction is even going on.

For Vrousalis's pit, we simply expand the concept of information to include the relative physical positions of the actors themselves. This isn't a big leap, since information is never not derived from the configuration of physical space. The person at the top of the pit possesses the fact that they are at the top of the pit, and the person at the bottom does not.

Domination, Fairness and Pareto Optimality

The subject of domination came up, and was rightly distinguished from exploitation, using the example of a school bully. The bully can beat up other kids and use the spectacle to dominate the rest of them, but it's only exploitation if he extorts their lunch money. Furthermore, if you define domination as the speakers did, as the power to intervene arbitrarily, you can have exploitation without domination in isolated scenarios of information asymmetry, irrespective of whether either party possesses such power.

And this isn't even touching oppression, which is precisely about using an existing power gap to widen it. Unlike exploitation, it's arguably only meaningful if the oppressor is deliberate. I realize I'll probably pick some fights with that distinction, but I don't see how you can oppress somebody without knowing you were doing it. You can certainly suppress people, and cause them real material harm, irrespective of gains to yourself, just by being clumsy, self-absorbed and inobservant, but I feel it's important to keep the concept of oppression tied to the bastards who do it on purpose.

I think we can agree that the concept of exploitation is tied to the enjoyment of material benefits, and furthermore in some kind of unfair or unjust proportion. I was tempted to introduce the notion of Pareto optimality into my definition, which is an allocation of resources which need not be completely equal, but helps everybody involved without causing anybody harm. However—to continue our cryptographic subtheme—we can imagine a situation where Alice either benefits directly at Bob's expense, or somehow captures part or all of Bob's share of their joint enterprise. As such, a transfer from Alice to Bob to make things fair would necessarily harm Alice and not be Pareto-optimal.

Of course, if Bob could carry out the enterprise in question without involving Alice, he would reap all the benefits himself, and Alice would have no claim on them whatsoever. If there is no arrangement between Alice and Bob, then there is no Pareto adjustment to make.

Back to Information Asymmetry

I'm pretty confident that information asymmetry is the key to a modern definition of exploitation. It isn't rights, because we can have exploitation without an infringement of rights. It isn't vulnerability, because vulnerability is implicit in exploitation. It isn't intent, because in the style of Dunning-Kruger, you can have asymmetric information without having the information that the information you do have is asymmetric. In other words, Alice doesn't know that Bob doesn't know what Alice knows, and just thinks he's really generous. Therefore:

Exploitation is the use of asymmetric information, intentional or otherwise, to initiate or prolong the capture of material benefit, that would otherwise be left to rest or go to other people, that would not be possible without said asymmetry.

In other words, Alice couldn't get the same terms from Bob if Bob had the information Alice had. The information in this case could be anything, from knowledge of the resale value of an antique, to having certain politicians on speed-dial, to being at the top versus the bottom of a pit.

This definition of exploitation also comes closer in range to the one used to discuss natural resources, and we all know how well that turns out. Using a definition of exploitation grounded in information asymmetry, we can all say something like this: Maybe I have information other people don't, and am getting a disproportionate share of the benefits in this situation, and they're getting the shaft. Of course, our culture's practice of professional specialization is designed to do precisely this: exploit unto others as they are likely to exploit unto you.

That's why I'm interested in the topic of exploitation. I'm seeing increasingly fervent protection of increasingly leaky silos of asymmetric information in a world that is continually becoming less and less able to support the lack of crosstalk. At the same time, we're into a second full generation of people who can't do basic things like change a flat tire or even feed themselves, as well as an increase in black and grey-market economies. In other words, we have all these problems and nobody can solve them because it isn't their department. There is likewise a profusion of people who are utterly helpless except for the one skill they have or property they own, out of which they squeeze every last penny. They are no match for the growing number of dangerously competent débrouillards, which is what you become if you survive being disenfranchised from that hog trough. Simply put, Mad Max will not honour your patent.

I sincerely believe we're witnessing the death rattle of a mode of social order that dates back to the time of Descartes. To be sure, institutionalized rent-seeking is guaranteed to be older, as is economic specialization, but the 16th and 17th centuries got the ball rolling on both practices becoming civilized and democratized—with further stiff shoves in the 18th, 19th and 20th. Now we're a decade and a half into the 21st century and we've already exhausted the utopic scenario where we all become elite professionals and politely bilk each other back and forth. One problem: not everybody can be an elite professional.

Co-exploitation by intermediating asymmetric information is also highly contingent on the ability to keep that information asymmetric. It should be no surprise that the first wave of giant internet businesses was all about disintermediation. It's disappointing, however, that the second wave is all about building new silos back up. Maybe not so smart.

To paraphrase Gilmore's Law, a networked society treats exploitation as a business opportunity and disintermediates it. The precarious thing to do when such an enterprise is successful, however, is to continue to exploit. As Bruce Sterling is fond of saying, those who live by disruption, die by it. The mid-90s were an anomaly, there won't be another one like it for a long, long time. But the genie is out of the bottle, and the ramparts keeping information asymmetric are just getting leakier. The only remaining civilized tool to preserve this state of affairs is the law, which has already begun: reverse-engineering your own personal property is now considered a crime. When that strategy inevitably fails, the only one left will be violence. I hear drones overhead.

Exploitation is an enormously important concept going into the 21st century, distinct from concepts like domination and oppression. It's important to generate a common understanding of what it is, because if we don't start voluntarily winding down the exploitation, domination and oppression are precisely what we're going to get.