e. j. wang

On determination

The following begins a multi-part essay about how the mind moves from abstract contemplation to concrete action and the obstacles that it encounters. I would recommend that you begin with my post on theorizing human decision if you haven’t read it yet.

1a. Startup

Between the idea
And the reality
Between the motion
And the act
Falls the Shadow

For Thine is the Kingdom

—T.S. Eliot, The Hollow Men

Everyone in Silicon Valley knows that it is hardly any longer true that a geeky Stanford student can build a large and successful company with just JavaScript, linear algebra, and a vague but firm conviction that he is doing the right thing. But it is worth probing into why this is the case.

By Silicon Valley standards, a successful company must capture, and optionally create, a large amount of value. To capture value, it must insert itself into a process that is valuable to someone. The value it captures is proportional to how valuable this process is to that someone. Hence, for the company to capture a large amount of value, it must embed itself into a process that is either very valuable to a few people or at least a little valuable to everyone.

A company embeds itself in a process (“acquires market share”) by seeking rent1. Rent can be sought in a number of ways; historically, the most acceptable has been Schumpeterian rent, or the arbitrage between local productivity improvements and a market in which these improvements have not become widespread. Though few would admit to it openly, most successful tech companies also seek another form of rent: network rent, which is the added value that a company can provide due to its preexisting market share. Network effects create natural monopolies; a perceptive observer may note that all honest network rents want (and tend) to become monopoly rents.

The function of the tech entrepreneur qua tech entrepreneur is to choose a problem and a solution to that problem. The success of a startup is determined by the value of its problem and the rents that can be sought with its solution. This is an underdetermined system. The rents that can be sought on the solution are either Schumpeterian rents, which are determined by the entrepreneur’s specific technical and managerial capacities, or network rents, which are determined by the entrepreneur’s relative ability to raise and sell.

Let us say that, for some reason or another, someone starts out with the abstract goal of starting a billion-dollar tech company — that is, of “making it.” Call this person the Aspirant. Regardless of what idea he chooses, he must become a good entrepreneur; that is, he must acquire the qualities required to execute well the function described above:

The Aspirant thus sets himself diligently about the task of acquiring these general, idea-agnostic skills. It is typically a good heuristic to make decisions as late as possible, and it is not yet time where the choice of idea has a meaningful bearing on the next action. In his education, the Aspirant acquires practical skills of how to act and think. He meets and befriends those around him who are worth knowing. He acquires, with a knowing irony, a number of impressive credentials; he cycles through a number of jobs and internships, soaking up experience and understanding.

But sooner or later, the time comes for action. And action is something different from contemplation. To enter the field of action, the Aspirant needs an idea. And based on this idea, he also needs:

The Aspirant can acquire all of these. But each requires a significant investment of time and effort — and, more importantly, they bear a severe opportunity cost. The capacities specific to one idea may generalize only marginally to others. Time spent on a bad idea is wasted. And the opportunity cost of spending time on a bad idea is as great as the number of alternatives the Aspirant has.

Stanford, my alma mater, is notorious for its startup culture. We are thought of in the popular imagination and in that of our humanities professors as a colony of entrepreneurial philistines, a “VC fund with a school attached.” But this “startup culture” does not in fact exist. Yes, Stanford has a serious infestation of techies. Yes, many of these techies want to start companies and strike it rich. They snag internships, publish papers, and network. They acquire everything they need to “do their own thing,” when the time comes.

But the time rarely comes. The opportunity cost is why.

1b. The undetermined class

Between the conception
And the creation
Between the emotion
And the response
Falls the Shadow

Life is very long

—T.S. Eliot, The Hollow Men

Why is it so hard for Stanford students to deal with opportunity cost? To answer this we need to ground ourselves in the concrete. Thus far, we have been discussing the Aspirant primarily in the language of abstract necessity. (Suppose someone were to optimize for this goal — then what?) But there are people who actually do optimize this goal, and through them it becomes clear what the very specific conditions are that give rise to it.

We have already posited that the Aspirant is probably male. Add to this that he is likely upper-middle-class, that he grew up in a comfortable suburb, and that he is either white or Asian. He attended an elite university; he spent much of his adolescence striving to get into this university, because he was raised under the understanding that it was the precondition for doing anything meaningful with his life.

The Aspirant is part of a larger class — call it the undetermined class — whose strategy from the start has been to maximize opportunity. An elite university is a great place to seek opportunity, as is an internship. For the second-generation immigrants of the undetermined class, this strategy is a multigenerational affair; America is nothing if not the land of opportunity. Choices are to be sought after, accumulated. Only then, after all, can they be made.

The word used in the lingo is not maximizing opportunity but preserving optionality. Because one must first have an option to exercise it, it is characteristic of the undetermined class to base most of their decisions on the preservation of optionality. Creating optionality is spoken of too, but mostly by people who deal with it in financial contexts; as used in common parlance, the term refers to pre-existing possibilities that are only ruled out, never created.

Undergirding the concepts of opportunity and optionality is the fundamental concept of possibility. Possibility is never created, only destroyed. The space of possibility, like potential, is at its largest at the moment of one’s conception. And it contracts with every action one takes from the moment of one’s birth. As a life unfolds, the space of possibility shrinks to a point — a point finally reached on the deathbed. But as long as one still lives, one lives in a state of unfinalizability:

Nothing conclusive has yet taken place in the world, the ultimate word of the world and about the world has not yet been spoken, the world is open and free, everything is still in the future and will always be in the future.2

The undetermined class is the proper subject of liquid modernity. It is characterized by two things: being born into a world of enormous possibility and setting a goal of maintaining this possibility. You are part of it if you have no clue what you will be doing in ten years, and especially so if that comes as a relief. You are part of it if you yearn to be released from the meager world you have already created for yourself, from your obligations and commitments, from the impressions that you invariably make on others. You are part of it if you fantasize constantly about returning to your youth, about changing the past and being released from the passage of time.

Who falls outside the undetermined class? Third-world peasants, fishermen, factory workers — but also academics, the vieux riche, royalty (well, usually). More or less, those born into circumstances they cannot or will not change. Though distinct from both, the undetermined class fits more or less neatly within the professional-managerial class (PMC).

In America, the undetermined class is characterized by a deeply unfortunate case of indefinite optimism. Some would describe indefinite optimism as the complacent acceptance of growth. I think this misses the mark. Indefinite optimism — the revealed preference of a class that spends more time positioning itself to act than acting — is not a kind of optimism at all. It is an indicator of fear. As opportunity is palpably less abundant here than it was sixty years ago, more resources have to be spent on securing it and fewer are available to realize it. Thus the PMC gives birth to the undetermined class. Its collective processing power is diverted from the socially productive game of organizing production to the zero-sum game of jockeying for opportunities. Stagnation begets undetermination begets stagnation.

Such persons, without waiting to discover how their desires may be realized, dismiss that topic to save themselves the labor of deliberating about possibilities and impossibilities, assume their wish fulfilled, and proceed to work out the details in imagination, and take pleasure in portraying what they will do when it is realized, thus making still more idle a mind that is idle without that. 3

The existence of a large undetermined class is in some respects a useful thing. It may be for the best that we have a good number of highly educated workers who cluster around the sector of the greatest productivity growth, be it finance or tech. At least in the short term, it helps us allocate labor “optimally” and fluff up our GDP.

But we also see the adverse effects. The four power centers of the US (NY, SF, DC, LA) swell into imperial strongholds independent of the rest of the country; other critical, less glamorous sectors of the economy fall by the wayside; deindustrialization, more than automation, wreaks havoc all across the hinterland. The undetermined class is the object of a deep resentment from the determined classes — especially those determined into irrelevance by the closing of American factories. What have they gotten in return for this devastation? Between finance, tech, politics, and media, why has this expanding class produced so little?

Before we can answer these questions, we need to reach a final understanding about the nature of the undetermined class.

1c. Determination

Between the desire
And the spasm
Between the potency
And the existence
Between the essence
And the descent
Falls the Shadow

For Thine is the Kingdom

—T.S. Eliot, The Hollow Men

Having now constructed the phenomenon of the undetermined class, we see immediately that this phenomenon is already overdetermined. The territory we have traversed has already been mapped in several different ways by a number of relatively famous thinkers, some of whose ideas I have referenced by name. Why do we need another curmudgeon essay about elite risk aversion, about “preserving optionality,” about complacency, indefinite optimism, low conviction, liquid modernity? Why take such an odd route to conceptualize a conclusion that has already been reached?

To understand what I have been trying to accomplish here, let us summarize our progress up to this point. We began, somewhat arbitrarily, from the aspiring startup founder. From the necessary struggles of this founder qua founder we unearthed the existing reality of the undetermined class; moving from abstract to concrete, we outlined the basic features of this class as it currently exists, as well as a materialist account of its emergence. We can now use our understanding the founder’s historical situation to answer previously unanswerable questions about him. Most pertinently, I believe that the concept of undetermination allows us to make practical claims relating to how the founder should go about surmounting these obstacles, and that it enables these claims to a far greater extent than the other concepts mentioned.

These claims require a final, more difficult theoretical leap. So let us press on.

The undetermined class is composed of undetermined subjects. Let us call determination the property whose lack is the essence of the undetermined subject. We would like a theory of determination; we would like to understand what sort of thing determination is, so that we may acquire some of it for ourselves. To this end we should begin by asking: well, what sort of thing is its lack?

The undetermined subject is undetermined because it is not fixed. For the emergence of the undetermined subject it was necessary that many of the social and cultural factors that previously held it in place have been removed, at least in part. But freedom, although necessary, is not a sufficient condition. The undetermined subject only appears when it is unwilling to exercise the freedom presented to it. We can rearrange these terms: the undetermined subject is free, but it has no free will.

What does this actually mean?

By freedom I mean the space of possibilities open to being actualized. By will I mean the process by which this space of possibilities is actualized. Freedom is being; will is becoming. Freedom is essence; will is existence. The undetermined subject has an enormous amount of freedom. This freedom is purely abstract. As the scope of freedom expands in history, there comes a point where the limiting factor to free will is not freedom, but will4. And indeed, the undetermined subject is characterized precisely by an abundance of freedom and a corresponding paucity of will.

What we want is a theory of determination. Thus we are operating under the assumption that there is a more fundamental quality, determination, whose absence is the cause of undetermination. And undetermination arises when freedom exceeds its will. Thus the subject becomes more determined when its will increases or its freedom decreases.

Note that these correspond precisely to two dictionary definitions of determination:

  1. Firmness of purpose; resoluteness.
    (free determination; determination as a function of will, with freedom held constant)
  2. The controlling or deciding of something’s nature or outcome.
    (unfree determination; determination as a function of freedom, with will held constant)

And if undetermination is the gap between freedom and the will, determination is this same gap considered in the opposite direction. It is the movement from freedom to will. It is the bridge between contemplation and action. We see now that there are two types of undetermination: a free determination, where will expands to fill its given freedom, and an unfree determination, in which freedom contracts to fit around a given will.

If we seek determination and we are unwilling or unable to give up our freedom, the determination to which we aspire must be free determination. To move toward free determination from a position of preexisting freedom is to move toward free will. So what sort of thing is free will?

A free will is commonly understood as one whose decisions are not determined primarily by external factors. But this is a rather coarse way to put it. Since Newton we have known that no such thing really exists. Each will is a flow in the stream of causality; there is no unmoved mover.

We will sidestep the endless tangle of language games surrounding the question of whether free will is compatible with determinism. By means of complexity theory I argued in an earlier essay that will, randomness, and computation are three components of information entropy.

Determination, being the resolution of uncertainty, is a form of information. So we may use the tools of information theory to reason about determination, and to make practical claims about how it may be acquired.


  1. To those who object that companies who create new markets are hardly seeking rent on them, I would respond that they have an overly narrow definition of what a market is. Social networks like Facebook tap into a human need for connection that defines a market in itself. The fact that they have increased consumption in this market (beyond, say, phone lines or the postal service) speaks only to the fact that they have obtained more rents than was previously historically possible. What must be kept in mind is that the seeking of rents is not necessarily a bad thing; rent enables the promise of superprofits that is necessary to compensate the risk that goes into innovation. Whether a specific rent is justified in this sense is a separate question entirely. 

  2. Bakhtin, Problems of Dostoevsky’s Poetics, trans. Emerson 

  3. Plato, Republic 458a, trans. Shorey 

  4. It is somewhat telling that liberalism as a political ideology has chosen this moment in history to run out of steam.