Maxime du droit civil

The academic year is almost over, as I finish up grading papers for this term. I'm now already starting to plan my teaching for the summer and next year.  I will be offering again my 112B - Berkeley's History of Early Modern Political Thought course - and I've started the process of ordering textbooks.  I am rethinking some of the textbook choices I've made in the past for this course.

Rousseau is one example.  In the past, I've assigned the Cambridge edition.  I can no longer do so.  There are serious flaws in the translation concerning the French term, droit.  This might seem like a petty point, but it actually is a big deal for me because it is arguably the central concept in the whole of the Social Contract which, it's worth remembering, is a work on principes du droit politique.

My Rousseau, Du Contract Social (Amsterdam: Marc Michel Rey, 1762), which I acquired earlier this year.   This is likely one of the early pirated versions based on the 12mo. 'third state' of M.M. Rey 1762, featuring a seated Justice.  I have identified this particular text as what Rousseau scholars refer to as the Dufour 136.

Droit is a central concept in this work.  English translations often render it as 'right.'

Consider the passage:


Rousseau, Social Contract (Cambridge, 1997), p.51 

This is a crucial point in Rousseau's theory of sovereignty and obligation, by taking on the question whether democratic self-binding is conceptually possible.  The analysis isn't all that original, and we get a clue concerning Rousseau's source of inspiration in the 'maxim of civil right' that he references here.

But what exactly is a 'maxim of civil right'?  The original French text expresses it this way:
Rousseau 1762, p. 21.

'La maxime du droit civil' isn't a reference to civil rights (that hasn't stopped students of political philosophy in seminars today puzzling over what a 'maxim of civil right' might mean).  Rather, Rousseau is talking about 'civil law' - that is, Roman law.  And what legal maxim does Rousseau have in mind?  This one:

D.45.1.108.1 = l.A Titio.§Nulla promissio.ff.De.verb.obl
.

This is a famous passage in the Digestum Novum of Roman law in the title concerning verbal obligations - stipulations.  The principle stated here becomes important in the intellectual history of sovereignty because it raises the question whether sovereigns (who are, by definition, legibus soluti - i.e., not bound by their laws) could nevertheless be bound by their contracts.  Bodin would cite this maxim in the République.

Bodin, République 1583: 132.

But one particular species of contract can never be binding:  A contract, or promise, due to oneself.  That is what this maxim addresses:  'No promise can be made that rests on the will of the promissory.'

This presents a special problem for theorists of popular sovereignty.  If 'the people' are both sovereign and subject, this legal maxim nullifies the force of any reciprocal obligation between sovereign and subject in a democracy.  To whom do citizens owe their political obligations?  A standard traditional answer is their sovereign.  But, as Rousseau just explained, the people are the sovereign.  Hence, the problem: Can the people be obliged to themselves?

Rousseau's solution, of course, is what leads him to formulate the concept of a general will, as distinct from any particular private will or even the aggregate will of all.

This is just one example of many where careless translation choices fail to convey a very simple, straightforward idea and has, consequently, done real damage to a generation of students in political theory.   This is why, as I always advise my students, it is vital to consult the original sources: Translations are never original sources.

  

Popular Posts