Seeing Like “Seeing Like a State”

by Henry Farrell on February 5, 2008

My long “post”: from a couple of months ago on James Scott’s _Seeing Like a State_ and Brad DeLong’s review of it enjoyed a temporary revival when Brad “republished”: it in his ‘DeLong Smackdown’ series. But I got a bit of grief from one reader, who thought that I had given Scott far too easy a ride. Which is probably true – while I admire the book, I do have many disagreements with it, which I would have gotten into if I had been reviewing the book proper, rather than arguing against Brad’s interpretation. One such disagreement popped up when I was reading it again for class a couple of weeks ago, together with John Brewer’s _The Sinews of Power._1

In one of the early chapters of _Seeing Like a State,_ Scott gives us a Foucauldian view of how formal knowledge increases the power of the state.

Legibility implies a viewer whose place is central and whose vision is synoptic. State simplifications of the kind we have examined are designed to provide authorities with a schematic view of their society, a view not afforded to those without authority. Rather like U.S. highway patrolmen wearing mirrored sunglasses, the authorities enjoy a quasi-monopolistic picture of selected aspects of the whole society. This privileged vantage point is typical of all institutional settings where command and control of complex human activities is paramount. The monastery, the barracks, the factory floor, and the administrative bureaucracy (private or public) exercise many statelike functions and often mimic its information structure as well. (Scott, p. 79)

NB that this isn’t one of the later chapters that deals with the specific phenomenon of ‘high modernism’ – as I read it, this is a generalized claim about how increases in legibility enhance the power of the state (or of whoever the eye-in-the-pyramid/dude at the center of the Panopticon) is. But it’s also _wrong._ There’s no necessary reason to believe that legibility implies a central viewer with a synoptic vision, or that it enhances the power of those with authority vis-a-vis those who don’t have it, or at least, if there is, Scott doesn’t tell us what it is. Rather, he assumes it. And here, John Brewer’s _The Sinews of Power,_ an account of the development of the British state in the seventeenth and eighteenth centuries, is very helpful in illustrating the things that _Seeing Like a State_ doesn’t see.

Brewer’s account doesn’t completely undermine Scott’s. After all, Brewer’s main theme is the importance of state fiscal institutions to Britain’s power capabilities over the relevant period. The British excise office had extraordinary powers of inspection and control – chandlers and soapmakers couldn’t “legally ply their trade without first being inspected by the [excise] officer who kept their vats, moulds and utensils under lock and key.”(Brewer, p.215). The government was also the most important actor by far gathering quantitative information. Manufacturers tried to stop the government from taxing their commodities by withholding details of how they were produced; the leather trade for example did this to stop any leather duty being imposed until 1697, when a rogue London leather dealer helped the excise service devise a new tax, and found himself driven out of business for his efforts. All of this is what Scott would predict – the state seeking to create quantifiable and systematizable information, and affected social parties trying to stop it.

But what Scott’s argument doesn’t capture is that quantifiable information constrains the state as well as enabling it. Brewer describes how lobbyists created a new public sphere, using statistics to argue for their own particular notions of the public good. As government departments could usually produce statistics to support proposed measures, lobbyists started to gather their own statistics, drawing not only on their own sources, but on governmental statistics too. Wire drawers, for example, argued for the repeal of the gold and silver duty by drawing on poor law statistics. Initially, lobbyists relied on private contacts within government for this information; later, they began more and more to use parliament. Thus, it was precisely the creation of a government statistics, together with the broader creation of a mathematically literate sphere of public debate (which itself was shaped in important ways by the state) that allowed private actors to universalize their specific interests and to argue against further state intrusion.

Dealing with the state became one important way in which specifically defined groups sought to gain advantages over their rivals. There was … no homogeneous business interest. There were commercial and industrial interests whose very variety owed much to the government’s imposition of a complex system of regulation. The state’s intrusion upon civil society created sophisticated ‘interests’ whose political conduct was, in turn, informed by the open and accountable political system in which they operated.(Brewer, p.249)

Brewer’s lesson, then, is that the politics of legibility are much more complicated than Scott’s focus on state and hierarchy might suggest. Sometimes, formal knowledge will indeed enhance the power of the central observer, the authority gazing down on its society. But there is no _necessary_ reason why this should be so. Sometimes legibility serves to constrain those central observers, and to increase the power of those at the periphery, by providing them with potent arguments that they can bring to a broader public sphere. Clearly, the existence of such a public sphere is a necessary condition for the benign uses of legibility. But in Brewer’s account, the actions of the state, in part unwittingly, gave rise to the creation of such a public sphere. Brewer implies that the invention of the English public sphere may have had as much to do with excise officials teaching trigonometry to local children as with arguments in coffee houses. As noted, this doesn’t mean that Scott is entirely wrong; far from it. But it does mean that his book should be read against other historical accounts to provide a proper picture of what is actually going on.

1 I assigned the books in the first week of my IR class – I wanted the students to get some sense of where states actually came from and how their fiscal powers got established. If anyone’s interested, the syllabus is available “here”:



Seth Edenbaum 02.05.08 at 10:10 pm

Ironic for DeLong at least that this post should follow
this one

I thought that “delivering people to the labour market” was the principal function of public transport rather than higher education.


trane 02.05.08 at 10:16 pm

“what Scott’s argument doesn’t capture is that quantifiable information constrains the state as well as enabling it”

I guess it is fair to say that the state is most often the villain in Scott’s work. But I do not think your reading above is fair. Scott points out at an early stage in Seeing Like a State that “the State is the vexed institution that is the source of both our freedoms and our unfreedoms” (something like that, at least, I don’t have the book at hand).

His main point is not that state formation is always bad, but exactly that states that are unconstrained by a public sphere, legislators unconstrained by constituencies, when guided by high modernism tend to go awry.

In his story, England would pass as one of the more successful (less bad) cases of state formation.


trane 02.05.08 at 10:20 pm

P.S. Your syllabus looks very interesting. It is a good point of departure for a course of that type to ask Where States Come From? But my impression is that IR courses seldom do so.


otto 02.05.08 at 10:37 pm

You might want to toss in a couple of chapters of Brian Downing’s “The Military Revolution” on, say, Prussia and Poland.


Henry 02.05.08 at 11:21 pm

Trane – I acknowledge that he does make these provisoes, but the statement that I quote seems pretty unambiguous and sweeping, and furthermore is about legibility in general, rather than in the specific context of High Modernism. I think that he could make a plausible counter-argument along the lines that you suggest, but this would require him to row back, I think on some of the claims that he makes about legibility, which pretty well all suggest that it leads to the aggrandisement of state power in the current mss (he does say that this isn’t all bad, but his argument seems to be that the reason why an increase in legibility isn’t always bad is b/c the state may be better than some of the brutal and domineering local elites who previously prevailed).

You’re right that both (a) IR courses should ask where states come from, and (b) most of them don’t. I also had a session that I wanted to teach on where international markets come from, with dollops of Greif, Polanyi (some of the less well known anthropological stuff) and others, but sadly it had to be cut b/c of lack of time in the semester (I am already asking the students to attend an extra session voluntarily if they can). Ideally, some day, I would love to do a complete course on this sort of stuff.

Otto – I haven’t read any of Downing – it sounds interesting. I have read Alf Luedtke on barrack society in Prussia which presented it as being pretty extraordinarily horrible.


Slocum 02.05.08 at 11:46 pm

“Sometimes, formal knowledge will indeed enhance the power of the central observer, the authority gazing down on its society. But there is no necessary reason why this should be so. Sometimes legibility serves to constrain those central observers, and to increase the power of those at the periphery, by providing them with potent arguments that they can bring to a broader public sphere.”

I’m reminded of an argument that Jerry Mander made about computers that struck me as wrong for just the same reasons. I can’t find the particular piece I was looking for but this is close enough:

People may edit their copy, communicate with their friends, connect with other like-minded people, and so on. But the computer doesn’t change the fact that great centralized institutions — corporations, trade bureaucracies, militaries, governments and so on — are able to use those same computers with far greater connections and with far greater real power.

What this misses is that “great centralized institutions” were able to gather and exploit information quite effectively even before the computer and internet ages. But widely dispersed ordinary people could not. So the effect of computer networking has been to reduce rather than expand the information advantages of centralized institutions. For example, it has long been the case that companies were aware of systematic failures in their products while their customers were ignorant. But now it is quite easy to discover that there are many other users experiencing the same failure you’re seeing.


mq 02.05.08 at 11:48 pm

Considering that the agent in IR is almost always the state, and the complexity and variation in state institutions, I find it astounding that IR courses don’t start by thinking clearly about what a state is and how it operates internally.


n4 02.06.08 at 12:13 am

Legibility in the examples given can cut both ways if you agree on the items or values that you are counting, and if you can find things to count. One of the problems with reforming health care or legal aid in the UK for example has been trying to agree what to count, trying to demonstrate the value of things that are hard to count, and the structural costs of counting.


Marichiweu 02.06.08 at 1:32 am

I recommend the festschrift (is that the word) to Scott published in American Anthropologist back in 2005. The anthropological uptake of Scott’s work is idiosyncratic but significant. The contributors here do a great job of exploring the strengths and weaknesses of Scott’s frameworks in a wide variety of ethnographic contexts.


geo 02.06.08 at 3:21 am

Perhaps this is off the subject — if so please ignore it — but is it possible that the Panopticon has had something of a bad rap? After all, guard-on-inmate and inmate-on-inmate violence (especially male rape) is epidemic in many prisons. Mightn’t a little more surveillance actually help protect weaker or less popular inmates?


Jeff 02.06.08 at 5:16 am

Who best addresses the fact that the state has its own interests, but is (probably primarily?) the site of struggle between various interests? If IR is the projection of these dominant interests externally, then how do we get to unified versions of these “great centralized institutions”?


Z 02.06.08 at 10:59 am

Sometimes legibility serves to constrain those central observers, and to increase the power of those at the periphery, by providing them with potent arguments that they can bring to a broader public sphere.

Indeed, Gérard Noiriel argues in État, nation et immigration that citizens of the periphery of France (think of rural farmers far from Paris and speaking in vernacular) willingly joined the massive social apparatus of classification, conscription and accounting carried by central power at the end of the nineteenth century because they thought that would give them more power. The central power was too weak to constrain the periphery into providing enough informations to collect taxes and organize draft but citizens willingly provided them because the perceived advantages were greater (chief among them, access to public education for their children). Once registered, and thus submitted to military and fiscal control, citizens of the periphery started to engage in national politics, if only to get their fair share of the budget.

This always struck me as a very important lesson in state formation: successful new powerful public entities are those which could be those able to entice their formal members into real participation by offering them important perceived advantages. The EU is currently good at it with respect to member states, not so with respect to citizens.


Doug 02.06.08 at 12:16 pm

“Alf Luedtke on barrack society in Prussia which presented it as being pretty extraordinarily horrible”

Until you compare it to accounts of traditions in the Russian army…


seth edenbaum 02.06.08 at 4:00 pm

As implied in my first comment the bureaucratization of knowledge has as much to do with capital as the state itself. If the market is fundamental than curiosity outside the bounds of the market is not. So we get the military industrial complex, the academic industrial complex and others. And we get discussions of economics and IR as if economics and IR were the world and not merely lenses through which to see it : “If we don’t see it it must not exist.”

The rest of us are left banging our heads against the wall in despair.

Comments on this entry are closed.