Consent of the Surveilled

I've been interested in the question of governance under conditions of mass physical mobility for a while. The interest is partly selfish, since I am one of those people with a romantic longing for a nomadic lifestyle. But now, there are better reasons to ask the question. In 2012, for the first time in history, there were over a billion international tourist arrivals worldwide. Chinese tourists led the way, spending $100B of a market of over a trillion dollars. The data isn't in yet, but it seems like 2013 might turn out to have been another record-breaking year. And that's just the beginning. What has started as a tourism boom is likely to end as a secular lifestyle shift enabled by mobile digital technologies. In a few decades, we might be living in a world where at any given time, only half the nominal population of a country is actually living and working in that country. A world with far fewer "vacations" but a lot more (and more extended) travel. At least, I hope that's the direction we're headed. Mobility, especially across jurisdictional boundaries, both domestic and international, is a problem for governments because it interrupts or complicates their ability to govern. This is why the forced settlement of illegible nomadic peoples is an essential part of any serious history of governance. As Julius Caesar once said, "hold still dammit, so I can see and rule you!" But thanks to surveillance technologies --  and this is the silver lining to the Snowden affair -- soon we might not need to hold still. Those of us who want to might be able to become nomads without dropping out of society. When a few fringe minorities such as gypsies are globally mobile, governments experience it as a tolerable annoyance that can be contained, like a nagging cold. When the travel bug turns into an epidemic, and populations mobilize en masse, governments are likely to experience an existential crisis. This is why I am not overly concerned about the rise of surveillance states in the long term. I suspect population mobility will catch up with, and possibly overtake, surveillance capabilities, and the social contract will be a pretty sweet deal once again. We'll accept more surveillance in exchange for more physical mobility. Virtual mobility online, which is already pretty unrestricted in liberal democracies, will not be as important in the renegotiation I suspect, but for what it is worth, it too is being actively renegotiated in the form of the Net Neutrality debate, among other things. So it is a good thing that the social contract is being actively renegotiated along digital lines worldwide, via the surveillance debates. Because governance, rather than work/life blending technology, is the bottleneck  in increasing physical mobility for humans today. As a personal example, I recently signed up for the TSA pre-check program, and I think the increased level of surveillance is totally worth it in return for more pleasant air travel. Agriculture may have immobilized humanity, but what is keeping us under-mobile today is the inability of governments to track moving objects competently. Not crops that need cultivating, machines that need monitoring or cubicles that need occupying. So a crucial question for the future of the mobile revolution is whether governments can get better at tracking moving governance targets. Surveillance technologies provide the means, but not necessarily the skills or principles. While governance skills are catching up to available capabilities, we must expect a good deal of incompetence (malice at the level of individuals and agencies can usually be attributed to incompetence at other levels, so in the larger scheme of things Hanlon's Razor prevails and it is all incompetence). Whether we can (or even should) get governments to stop tracking us altogether is, in my opinion, a different and much less serious question, because the answer is almost certainly "No" (and "No"). The answer to the competence question has to do with the basic needs of governments. What Governments Want How governments see is a complex question, but one for which there is a surprisingly good answer. What governments want is a more difficult question that has no such good answer as yet. To answer it, you have to start by recognizing that governments have a unique corporate nature (usually embodied by a bureaucratic civil service)  that makes them different from other corporate entities such as individual citizens, businesses and non-profits. While they share self-perpetuation as their basic drive with other biological and artificial entities, what makes (organizationally embodied) governments different is that their basic survival needs are different. We like to pretend that the basic needs of government derive from the desire to fulfill the nominal social contract and current mandate of elected representatives, which we believe confers legitimacy on the existence of governments. We like to pretend that governments ought to exist solely to serve us. This is seeing like a citizen, the flip side of seeing like a state. A view of the state as, ideally, something like the bovine creature in Douglas Adams' Restaurant at the End of the Universe, which had been carefully bred to want to be eaten. The social contract isn't a contract between one kind of entity and another kind that exists solely to serve it. It is between two distinct kinds of entity with partially conflicting intrinsic needs. Needs that drive a non-trivial ongoing renegotiation of the contract at all times. The faster the pace of technological evolution, the more active the perennial renegotiation. The only asymmetry is that one side cannot help seeing the other in limited ways, while the other chooses to do so. So citizens believing that states exist to serve them is as misguided as states believing that citizens and businesses exist to serve them. Except that citizens voluntarily choose that belief (a choice characteristic of incompetent parents and managers) while states are necessarily constrained by it. As with any negotiation, to understand the current state of highly active renegotiation of the social contract, we need to understand the needs of the other party. We need to understand the unique way a state satisfies its self-perpetuation drive, driven by its constrained ability to see and understand what it is doing. We need to appreciate what it means to struggle for survival like a state, fumbling and stumbling about with limited intelligence in semi-darkness, dimly aware that our existence depends on our ability to satisfy the capricious needs of entities called citizens. Surviving Like a State Governments are agents too, but they don't seek life, liberty and happiness as citizens do. They don't seek to maximize shareholder value, serve customers or serve a core group either. All the complex needs of governments arise from three basic ones: counting, conscripting and taxing the citizenry. This is what governments do to survive in their day-to-day lives as artificial organisms, just as humans eat and sleep and corporations buy and sell, in order to live another day. When we speak of consent of the governed from an everyday perspective, we don't really mean an implied contract based on ceding a monopoly on legitimate violence to the state, in exchange for some security. That's too abstract. What we actually consent to at an everyday level is being counted, potentially conscripted and being taxed through specific mechanisms such as periodically showing up at the DMV, to renew our drivers license, a point that the creators of South Park instinctively understand.

1. Counting (more generally, tracking), at any level from the simplest once-a-decade census to brain-implanted chips wired to the NSA's thought-monitoring quantum computers, is basic. Without knowing the size and social/cultural structure of the governed population, and being able to effectively locate and communicate with any specific individual, a state cannot survive anymore than humans can without being able to tell poisonous and edible berries apart.
2. Conscription is equally basic because a monopoly on legitimate violence is useless without the ability to recruit some citizens to exercise it against others, or against foreign powers, through specific organs of state (military and police forces and more subtle kinds of violence through non-physical weapons such as central banks).
3. Taxation, being the raison d'être of all governments from pure kleptocracies to pure welfare states, is basic as well, and is embodied by organs of state devoted to collection, redistribution and regulation.

It does not matter whether these needs are pursued in ways that align more or less with particular special interest groups through mechanisms ranging from legitimate influence to regulatory capture (the various x-ocracies that can in practice lend a personality to governance). Who the state works for has little to do with how the state works as an organism. These three basic needs, when satisfied, allow governments to self perpetuate, and in the case of liberal democracies, peacefully procreate. They constitute the bottom layer of the Leviathan's Maslow pyramid. They will not go away as long as we need governments for any purpose at all. Any sort of governance apparatus we build for ourselves or have imposed on us, from local muscle enforcing a protection racket for the reigning bandit, to the NSA snooping in your inbox, will automatically develop these basic needs and evolve mechanisms to fulfill them. This is because ceding any sort of meaningful agency is equivalent to creating real agents whose personalities are defined by what is ceded. When we cede the capacity for violence for instance, we create agents of violence that count, conscript and tax through the sword. A sword is the most primitive sort of technology of consent. A technology of consent is how a state meets its own basic needs in order to survive. From the point of view of the governed, the name is appropriate because it determines how consent is obtained and the nature of the compensation delivered in return. Technologies of Consent From the point of view of citizens with more complex models of themselves than the state can understand, the biggest component of consent is in fact consenting to be seen in a limited way. Despite our suspicion of surveillance, in general we prefer to be seen as complete, sum-greater-than-parts individuals (i.e. recognized in a philosophical sense) rather than collages of narrow identifiers. But the crucial thing to understand is that the biggest element of what we sacrifice is completely invisible and useless to the state. So entering into any sort of consent negotiation expecting this sacrifice to be acknowledged, appreciated and somehow rewarded is to misunderstand the nature of the beast. Even the most benevolent state cannot compensate you for what it cannot see that you have sacrificed. Individual politicians and government functionaries might, as humans, acknowledge it, but not the state qua state itself. As with relating to a child or a less developed adult, we must enter into negotiations by accepting that we will be seen in a limited way that falls laughably short of philosophical recognition. And we accept the risks that accompany being viewed as less than we are, by an entity with more power over us than we ourselves possess. Risks that are similar to the ones engendered by the Dunning-Kruger effect, but on a massive and dangerous scale. Fortunately the prevailing technology of consent is actually evolving so we have to make less of a sacrifice. There is a good chance states can evolve to keep track of a highly mobile, but still meat-based citizenry without significant institutional change within governments. And without ultimately ineffective moves like brute force computation in service of a bureacratic panopticon that can see everything but understand nothing. The surveillance state in its mature form might even be a good thing if it gets sufficiently smart at understanding what it sees. The idea that consent is obtained and rewarded through a specific technology rather than an abstract principle has a subtle but important consequence. Since states cannot see at all except through a technology of consent, from the point of view of the state, the consent is being obtained from entities created by the very technologies used to obtain it. This is the medium-is-the-message effect in governance, and causes governance to co-evolve with technology. The key to understanding why this evolution might be positive is to think of the social contract renegotiation itself as being mediated by evolving technologies of consent. This is like two beginners negotiating who gets to fly a plane after accidentally taking off. When the prevailing technology of consent is forced to change due to technological evolution (and a resultant change in the behaviors of the governed), governments experience internal anarchy and exist as Hobbesian homunculi within their societies, during the transition period. In other words, governments will internalize digital technologies as a new technology of consent, and will remain in flux until this process is complete. During this period of flux, citizens have more leverage, not less,  because they can learn faster than the apparatus of the state can. Let's look at this in more detail.   The Evolution of Consent When the technology is swords, consent of the governed is realized as consent of the intimidated.  When the technology is land titles, it is realized as consent of the settled. When the technology is paper census forms, it is realized as consent of the surveyed.  When the technology is IQ tests for military mobilization, it is realized as consent of the intelligence-tested. In between stable technologies of consent, the state has no formal idea what it is doing and blunders about blindly. During that period of blindness, citizens have a window of opportunity to influence the shape of future governance. So the evolution of consent can be viewed as the the changing levels of legibility experienced by a state, and actions unilaterally undertaken by citizens during its periods of blindness. When the state is more blind, it has no idea what citizens are consenting to and what it needs to offer in return. When the state is less blind, the contract gets clearer to both sides. Technologies of consent need not be explicitly designed around the basic needs of governance (counting, conscription and taxation). The needs of governance might be fulfilled as a side effect of fulfilling some other need, including a need on the part of citizens. In fact, that's probably the better, and more common model (and the reason national identity card schemes with unspecified purposes might be a bad idea). Authoritarian high modernism could probably be defined as the design of governance mechanisms driven by governance needs. The opposite sort of state is one that seeks to satisfy its own needs as a side effect of satisfying the needs of the governed. When that happens, you manage to fulfill two distinct needs on both sides of the social contract with a single, historically specific (and therefore, self-limiting in time and scope) technology of consent. You satisfy counting, conscription and taxation needs on one side, and obtain consent and deliver value in return, in a historically and geographically specific way on the other. It is a different way of approaching the problem of achieving limited government, through the design of governance in the form of mechanisms that are designed for obsolescence. There is no need to worry about laws accumulating faster than they are repealed if they fade into irrelevance naturally. The constraint is that the immediate need co-opted as a technology of consent has to be nearly universal. So for a car-based nation, consent of the governed translates to consent of the licensed-to-drive.  The United States Government (USG) sees its citizens primarily as entities-that-drive-around, and secondarily as entities-that-need-Social-Security-in-retirement. For a nation based on rationing of basic commodities, such as India, consent of the governed translates to consent of the underfed. Today, middle-class India has no real need for the Public Distribution System, but still needs ration cards to deal with the government for any purpose, because the Government of India (GoI) still sees its citizens primarily as mouths-to-feed-cheaply. It is an obsolete technology of consent, but in my view, wiser than a generic national identity card. If the sine qua non of the United States Government turns out to be healthcare delivery in the future, then consent of the governed will translate to consent of the unhealthy.  I hope this doesn't happen. Anecdote: earlier this week, my wife and I finally manged to sign up for Obamacare. The reason our application was causing errors was that neither of us was in any existing federal database that delivers any sort of welfare. Technically, Obamacare appears to have been built as an extension and integration of existing services that target the disadvantaged, so parts of the US government literally cannot see citizens who have never relied on the welfare arm of the state for basic needs (another roadblock was that the system had not yet learned to see me as a citizen). Consent of the Mobile Hopefully, healthcare delivery will not come to define consent of the governed. In the best-case scenario, geographic mobility will. The reason this is the best case scenario is that increased mobility, unlike better healthcare, probably means more all around economic growth. The glass-half-full version of consent of the surveiled is consent of the mobile. Under conditions of mass mobility in lifestyles (online and offline), consent of the governed translates to consent of the surveiled. So far, there is no codification of this consent in a government-provisioned medium. The Hobbesian anarchy within governments has not yet settled. We rely on some mix of passports, registration on government sites, email addresses, regular net/Darknet distinctions, drivers' licenses, business licenses and social security cards to govern mobility online and offline. There is good reason to be optimistic that this soup will yield some sort of workable technology of consent upon which to base the future of governance: mobility has been the basis for the most recent, liberating and economically fertile technologies of consent in history: passports, visas and drivers' licenses. So the key to restoring the balance of power between governments and the governed is not uncomprehending fear of surveillance, but understanding the mobility-surveillance trade-off equation, and figuring out how much increased mobility we can demand in return for consenting to increased surveillance. Once we figure out how to increase our mobility to match the limits of the new technologies of consent, the trade implied by the new social contract will be fair once again. If governments can now see us better, we should expect and demand significantly increased mobility. Governments that can see better should have less need to limit legitimate mobility to fulfill their own needs. In fact, across the world, we should hope for relationships with governments that are defined by mobility. If, as I hope, surveillance (rather than healthcare delivery or something else) becomes the technology of consent, society will evolve to maximize mobility up to the limit of surveillance capabilities. States which recognize that mobility is economic destiny, the way geography once was, will prosper. States that limit it will risk economic failure. What does the prosperous case look like? Governing Mobile Populations Let's speculate a bit here, assuming that digital connectivity becomes sufficiently pervasive and global (which seems inevitable at this point). Just how physically mobile can a citizenry become, while still allowing relatively unchanged governance models to fulfill their core functions by learning to be effective surveillance states? What does effective governance with the consent of the surveiled look like? We only need to look at the three basic functions. If those can be fulfilled under conditions of maximal mass mobility, everything else should be possible.

1. Counting should be easy. So long as there is a nominal record of nearly all individuals based on a unique way to contact every individual (such as a registered cellphone or some sort of verified email address), and a way to keep that contact information current (such as requiring periodic cellphone verification), it should be possible to keep an accurate and current view no matter where in the world a person wanders. If the state actually provides some geographically portable benefits with sufficiently high frequency, such as a basic income or healthcare, that should be enough of a carrot to track nearly everybody. If not, a stick will be required, such as more frequent (but less painful) license renewal requirements, or some sort of meatspace verification mechanism modeled on parole rather than prison (perhaps in-person check-ins at DMVs in place of proof-of-address requirements to conduct any sort of business). No significant institutional changes are necessary. Just more competence at buying and operating software infrastructure. Models of constructs such as household may need to be updated, but there is nothing making counting functions fundamentally harder.
2. Conscription is a harder capability to maintain with a mobile populace, but is increasingly irrelevant in a world of automated warfare waged by small and highly professionalized military and police forces. Even if (say) half the population of America happens to be traveling in other parts of the world at any given time, that should not significantly affect the ability of the American government to wage war on some unfortunate little country. Again, no significant institutional changes are necessary. Since governments will most likely only need to conscript individual specialists in the future (such as say the top drone hacker in 2030, call him D'Rambo, who happens to be temporarily living in a Thai Buddhist, monastery fixing human-driven cars), conscription transforms from a problem of large-scale logistics to special-ops find-and-recruit logistics.
3. Taxation as you might expect, is the actual difficult problem (along with its alter-ego, regulation). If you've ever moved states within the US during a tax year, or worked significant periods in different states while being based in one, you know what I am talking about. Governments tend to be so bad at efficiently taxing mobile populations (and corporations) that citizens and businesses limit how they move merely to simplify their tax lives. Not all kinds of taxation are equally unfriendly to mobility. In the US for example, federal income and sales taxes do not limit national mobility at all. State taxes do. City taxes may or may not, depending on whether or not they over-reach into burdensome licensing models. Corporate taxes follow the same rough pattern. When you get to international mobility, taxation is messy enough that it can be a show-stopper.

This suggests that if governments need to change at all at an institutional level, the required changes involve the taxation apparatus. Or more generally, the taxation-regulation-redistribution apparatus. From an American perspective, the bottleneck governance level for taxation is the provincial one, between national and city governments in size, scope and complexity. This is not a novel observation. Many who think about the future of governance seem to have converged on the idea that the middle is the bottleneck level. Province-level taxation (as well as licensing and regulation) has already served as the scene of early battles between Internet-era businesses that try to serve a highly mobile population (such as Amazon) and governance systems that assume a highly sedentary one. Market in the Middle I think the most likely end-game here is the replacement of provincial governance institutions in their current form with a set of functionally equivalent market-like (but not free market) mechanisms. I call this a market-in-the-middle model. It is different from both the models most commonly discussed today: highly decentralized city-state polities and "barbell" polities with strong national and urban governance but weak or missing provincial governance. Market-in-the-middle assumes provincial governance will get reorganized along non-free market lines rather than eliminated or weakened. Local and national institutions probably cannot come to resemble marketplaces for a variety of reasons, and will likely remain corporate in form, but I see no reason the middle cannot be more like eBay than GM. An early example of this can be found in how road tolls are starting to be assessed. In Washington, tolls for crossing some bridges are assessed by photographing vehicle license plates and mailing bills to registered owners, irrespective of where they live. Presumably this sort of thing is achieved by coordination among DMVs of different states. There is no reason such mechanisms cannot be expanded to cover every sort of cross-state transaction, especially those that don't involve terminii at the individual or business level (pure middleware functions).  Similar mechanisms already exist in other functional areas, such as water rights management, coordination between state and region level electricity grids, and so on. I'd happily consent to being automatically tracked at airports, ports and highway state borders if I don't have to file income taxes in every state, or keep changing my drivers license. The specific technology of consent that provides the user experience of citizenship will likely be be driven by the internal architecture of the middle-markets that emerge. Purely as an example, imagine all dealings with all levels of governance between city and nation being mediated by the (already portable) area code of your cellphone. You pay all your taxes to the state that contains your area code. Everything else happens in the background no matter how you move around, via some hidden book-keeping based on border crossings. Or for a wilder example, assume that the primary compensation for surveillance is a basic minimum income (or better still, a negative income tax). How might that be delivered? How about you get paid each time you experience a tracking event, such as being photographed or tracked online by the government, with the payment depending on context. The more mobile you are, the more you get paid. These specific realizations of surveillance as a technology of consent need not be the ones that we actually end up with. So long as reconciliation of tax liabilities and other province-level obligations in different geographies happens in the background with little or no need for citizens to do anything, the needs of a mobile population are served. If this seems like a very banal conclusion to draw from a very impassioned debate, that's a good thing. The evil and good that states do is a matter of a balance of banalities.