Law in the Internet Society

Code is more than law

-- By KalmGustav - 13 Dec 2012

Computer programmes and law as coded media

In a very clever article Cornelia Visman and Markus Krajewski draw parallels between the codification of programming languages and the codification of law. Both law and computers (especially ever since the internet became widespread) are system of transferring communication. For Visman and Krajewski it is problematic: “law that claims to be able to regulate a technological medium merely produces a paradox, because it aims to regulate nothing other than its own grounds of existence.” (p. 91) In their understanding the law has lost its hegemony amongst the different communicative media and has now been imposed the logic of the computers.

But there is more that connects law and internet and makes them different from other media. In development of both law and computer programming, basic units of comment have been assembled and fixed in new units through codification. The first example of codification was the compilation of Roman jurists comments of cases into Digesta (a digest). The digest was in compiled in a Codex, the Latin for book, whence also the word ‘code’. Lawrence Lessig reminds us how the code of computer programmes is essentially like law – the law that determines the functioning of computer programmes. And like the law in many countries, the law of computers has been codified. The codification of computer programming commands into codes and protocols found place over the past decades when commands made by programmers were turned into code by programmers of code. Hence the codification in both and the appareant juridism of computers.

Meaning and interpretation of code

Allthough the parallel is tempting and helpful in some ways, it relies on an idealistic irrealistic vision of the law. The codification process for both law and Internet programming might have been similar and in both cases the foreclosure of debate seems to have found place. And this certainly is a problem.

Despite these similarities, the basic building material between the two is very different. The vision of law that Visman and Krajewski project is an idealized version of law that functions as a system. But legal rules are not stable and apply differently to different situations. There is room for interpretation. There have been debates to the range to which legal rules have a stable hard core or whether they are purely adaptive to the situation. Kelsen and Hart have both seen legal rules as having a strict core which is stable and foggy boarders wherein there is room for interpretation. Even this vision of the nature of legal rules is considered too optimistic to the stability of the norm by some. Duncan Kennedy has argued that both Hart and Kelsen are wrong in making such a distinction and that legal norms don’t even have this hard core.

Programming however is different for the mere fact there is no interpretation. A command or piece of code always results in the same result in binary code and hardware behaviour. The wording of a legal rule can stay the same over centuries, but the interpretation by law’s subjects, the courts and other law enforcement authorities can vary over time and radically change the meaning of a rule. The same is not true for a command in computer programming. A command as piece of code will still comprise the same directions for action for the chip. What will vary are the commands used depending on what type of applications are created for computers.

Codification of law is hence more supple. The rules even if codified are still open for interpretation and their meaning can change over time. Codification in computer programming however immutes the rules. The flexibility in computer programming over time will come from the addition of new commands to the code. It remains though that once a command exists, its meaning cannot be changed. As human society evolves and needs different applications for their computers to correspond to their needs this can become a political problem. The codification of computer programming is a political act, because it performs as the law of superstructure of the virtual space. At the same time there is no public control over it. This raises serious issues of democracy.


There is very little democratic control over the creation of code. This control however is highly necessary, because computer programmes affect human behaviour and thinking. However it seems far from our everyday lives and most people have little if any understanding of how computer programming works. One could say that computer programming will finally correspond to the desires of the public as its content will be dictated by demand. This might be a naive vision of the functioning of the marketplace for the desire and offer of new commands in programming. Past offer will detect future demand as the desires are structured by what exists previously. Such a situation hardly corresponds to democracy understood in light of Cornelius Castoriadis’s writing as a constant process of autonomously creating the institutions that govern us and not accepting external heteronomously created governing institutions. A solution is needed.

This essay draft is more compelling to someone who lives in a legal system that has been codified. Because you're not in one of those systems, your reader doesn't have the pun on "code" to provide some inherent reassurance that there is a point to the comparison. And of course, because this is also the home of legal realism as well as non-codification (which go together, because in a codified system no realist will ever become a law professor), your reader isn't likely to believe that law and computer programming have anything much to do with one another. This is particularly true, I think, if she knows a great deal about both of them and practices at the intersection. Then, in addition to believing that law is politics under another name and hasn't anything determinate about it, she will also know that law is about the way peoples' brains work and programming is about the way computers' brains work, and they're different. (This difference transpired because computers were made by people who thought they were making machines that imitated their brains, but that's a subject for another day.)

So I think I would revise by asking whether someone who actually knows about code is the appropriate reader for the essay. You're stuck with such a reader, and you might want to make the best of it.


If these would have been useful to people, why not link to them? If they are not useful to people, why are they here? In particular, are you sure you understood Duncan Kennedy correctly?

Cornelius Castoriadis, Domaines de l’homme, Seuil, Paris, 1986.

Duncan Kennedy, Legal Reasoning, Collected Essays, The Davies Book Publishers, Aurora CO, 2008.

Lawrence Lessig, Code Version 2.0, Basic Books, New York, 2006.

Cornelia Visman and Markus Krajewski, “Computer Juridisms”, Grey Room 29, Winter 2008, pp. 90–109.


Webs Webs

r3 - 23 Aug 2014 - 19:33:50 - EbenMoglen
This site is powered by the TWiki collaboration platform.
All material on this collaboration platform is the property of the contributing authors.
All material marked as authored by Eben Moglen is available under the license terms CC-BY-SA version 4.
Syndicate this site RSSATOM