Holmes on agency and contract 1891 4

Lawyers like to think of Holmes as a "legal realist"---a term that sounds off to a philosopher, I think. Part of this is verbal, but it seems more accurate to call him a "legal pragmatist": circumstantial evidence exists (he was a member of the Metaphysical Club with James and Peirce and the rest), but so does textual evidence (his emphasis on common-sense in judicial reasoning, his eschewing of abstract principle, and so on).

I guess I join Brian Leiter and others who've done a lot of work to call attention to this, though Leiter (among others) has spent effort to clear things up and help the "realist" out a bit (Leiter, 2001, "Legal realism and legal positivism reconsidered," Ethics 111: 278-301). A quick look at his paper pre-convinces me that he is right. (A longer read will settle my thinking for real.) The gist of my thinking is that Holmes can be said to be pragmatist as a matter of approach to legal reasoning, while Hart and the positivists are giving a theory of the nature of law itself. Leiter holds that view, but goes much further in showing how the two can actually depend on each other. I need more to persuade me on that point.

In any case, Holmes' approach to law is clearly common-sensical and unprincipled (not in any anarchic sense, but in an anti-positivistic sense). Why not call this pragmatist?

Now the important part of this point. Holmes effectively shuts down the two familiar moves in legal philosophy---that of the positivist and that of the naturalist. Is this a good thing? We'll see. (But the Leiter caveat from earlier: maybe the realist actually needs a version of the positivist's appeal to rules. This is getting a bit complicated and will need its own post later.)

The positivist---take Hart as a paradigm---can make an appeal to the rule structure that constitutes law, and argue that those rules (properly understood) both guide and enable the jurist's reasoning. But Holmes can't make this move given his rejection of such legal abstractions. (Could he respond that the rules of the positivist aren't necessarily problematic---perhaps they are not necessarily abstract? Hard to see how he makes that move.)

The appeal to morality and the common good---the sort the natural law theorist will want to make---doesn't seem available to Holmes either.

The urgency to figure this out comes in part from thinking about how legal philosophy ought to explain the slave's place in the republic. 

At a high theoretical level the question plays out as pressure to moralize the law in the face of an egregious set of statutes (slave codes, Jim Crow laws). MLK's famous civil disobedience arguments in Letter from a Birmingham Jail represent this (leftward?) pressure on the Holmes-style legal theorist. 

But at a lower theoretical level is Holmes' discussion of contract law and its origins in Roman slave-master relations. If masters took their slaves to be agents for the purpose of making contracts, then the obligations thereby created (a cornerstone of contract law) arise pre-conventionally---at least to the extent that the families that slaves joined are pre-conventional. If so, then Holmesian contract law is naturalistic, a seeming violation of his realism.

This kind of reading is complicated by Holmes' approving quotation of Hobbes, though. Here's Hobbes:

Likewise children, fools, and madmen that have no use of reason may be personated by guardians, or curators, but can be no authors during that time of any action done by them, longer than (when they shall recover the use of reason) they shall judge the same reasonable. Yet during the folly he that hath right of governing them may give authority to the guardian. But this again has no place but in a state civil, because before such estate there is no dominion of persons. (Hobbes, Leviathan, chapter 16; emphasis mine).

Here Hobbes claims that there is no dominion of persons in the state of nature, and hence (I think) ex persona domini powers are non-natural. If so, then Holmes can be taken as claiming that the slave-master origins of contractual obligations are conventional after all.

Have to figure this out. 

Holmes on agency and contract 1891 3

Questions raised by the slaves acting ex persona domini principle: 

1. That slaves personify masters---is that pre-contractual, or a consequence of the contract made between master as slave-purchaser and slave-seller? (By pre-contractual, I mean does it result from something other than the contract; I don't mean to suggest it's already operative prior to the slave purchase, since no ownership would exist yet.)
2. The slave enjoys an advantage of directness (family-member-like, unlike strangers). Is this a species of the idea that a man is head of household, the "representative" of the woman and the children?
3. Slaves complicate the legal philosopher's task of inventorying the short list of legal concepts---let's include three to start with: the concept of the author of an action, the concept of property, and the concept of contract. Slaves muddy all of these.

Holmes on agency and contract 1891 2

Notes on Oliver Wendell Holmes, "Agency" 5 Harv L Rev 1 (1891-1892).

Last post I briefly introduced Holmes' view that an agent has all and only those powers granted by the principal. He discusses an anomaly though which is instructive: "the rule that an undisclosed principal may sue or be sued on a contract made by an agent on his behalf" (1).

(A helpful discussion by Cristine Christodoulou is here; a law review article by Randy Barnett is here; an older paper by Grover Heyler is here; and of course Wikipedia is here.)

The oddball part of this is that an undisclosed principal can, in a sense, be party to a contract they didn't literally make (though their agent did). An undisclosed principal, aside from the havoc they represent in contract law, also runs counter to the philosophy of law, in that authority is natural (as I asserted in my previous post) but is conferred in virtue of the public recognition of that fact (as Holmes argues in his 1891). If a principal is undisclosed then there cannot be public recognition that they have conferred power on to an agent; hence, there was no such conferral, and the resulting contract is invalid.

So what to make of this anomaly of US and British law?

Let's first appreciate the bits of the philosophy of legal logic Holmes helps himself to in his discussion of public understanding (as the ostensive extension of powers).

1. Conferral is public and objectively accessible. The meaning of the powers is not impressionistic, mental, or private.
2. Ostension as pointing is pure referring and assigning; if it is mediated, it is the least mediated of all significations.
3. Extension as the objects referred to (that is, the agents granted this-or-that power).

The anomalous piece of law, observes Holmes, would "seem to follow very easily from the identification of agent and principal," a relationship resulting from the "power of contracting through others, natural as it seems, [that] started from the family relations, and that...has been expressed in the familiar language of identification" (2). The slave is the extension of the persona of the principal, growing from the family nexus in which slavery would have taken place in the Roman context. 

[B]y the Roman law contractual rights could not be acquired through free persons who were strangers to the family. But a slave derived a standing to accept a promise to his master ex persona domini. [ST--that is, slaves personify masters]...An obligation may be acquired through slaves or free agents in our power, if they take the contract in the name of their master. (2)

Not only did slaves thus have a privileged legal role in contract-making in ancient Rome (according to Holmes), but they did so in the context of their being identified with their master and their family. In a very real (and perhaps natural) legal sense the slave serves as a legal extension of the master. The transitivity of obligation thereby achieved suggests a domestic and intimate (albeit problematic) source of obligation in the natural state.

Holmes on agency and contract 1891 1

Notes on Oliver Wendell Holmes, "Agency" 5 Harv L Rev 1 (1891-1892).

"A man is not bound by his servant's contracts unless they are made on his behalf and by his authority, and that he should be bound then is plain common-sense." (1)

A pair of central concepts in the theory of legal obligation are that of [1] the authority of the individual acting as principal and [2] the transitivity of that authority once delegated to an individual acting as agent. 

Regarding [1]: this is a bit of natural theory, in that a person acting is the author of their act, independent of any institutional recognition of such authority; it is in that sense that it is natural, rather than conventional or institutional. One is author of their actions even in a state of nature. If that is so, then whatever obligation issues from authority does so even in a pre-institutional state of nature; thus, author-derived-obligation is natural.

Regarding [2]: it is less clear that this is natural, since transitivity is a logical property owing to the logical form of authority, and it is a matter of the philosophy of logic whether such a thing as "the logical form of authority" could be natural. Besides, something feels different once we enter a world of conferred obligations: the social dimension invites regulation (something like a framework of protection against abuse or whatever, or else something like a guarantee that a breach-of-contract proceeding could be carried out, etc.). But what exactly is it that "feels different" here?

I think it may be more than just the social as such. After all, the context of original authority is social as well, as critics of liberal legal and political theory are fond of pointing out. They are clearly right. We are born in families, we discover our egocentric selfhood even as we interact with caretakers and siblings and the rest, and we come into childhood and adolescence forming our character etc in relation to those around us. To hold that authority is natural for being non-social while holding that transferred authority is non-natural for being social would invite these well-founded criticisms.

I think it may be more due to the fact that we transfer authority via acts whose structure is largely linguistic---this is close to circular; I'll fix it as I go---and that are understood by the public that way. Says Holmes:

It is true that in determining how far authority extends, the question is of ostensible authority and not of secret order. But this merely illustrates the general rule which governs a man's responsibility for his acts throughout the law. If, under the circumstances known to him, the obvious consequence of the principal's own conduct in employing the agent is that the public understand him to have given the agent certain powers, he gives the agent those powers. And he gives them just as truly when he forbids their exercise as when he commands it. It seems always to have been recognized that an agent's ostensible powers were his real powers; and on the other hand it always has been the law that an agent could not bind his principal beyond the powers actually given in the sense above explained. (1)

Ostensible authority---authority conferred by explicit action or gesture (on the model of pointing)---is the determinative category, and not any merely intended conferring of authority. (No one can appeal to a mere intention to have so-and-so represent you in a contract negotiation, or else some sort of preference, or whatever. Legal obligation cannot be private and mental, but must be demonstrated and public.) But it doesn't stop there. If "the public" comes to form the belief that so-and-so has conferred authority on this one or that one, then they have. The public understanding of such a conferral is sufficient for that conferral to have become effective. 

But why should that be?

Here is where Holmes' pragmatism becomes very obvious. What other standard could the law use? If the issue is the intention of the one who has authority, then the law is contingent on a reading of something unknowable apart from the very party who is interested in the matter, and thus has a motivation to affect or alter it. That would obviously be a disaster. 

The problem though for this sort of pragmatist starting point is that there's some sort of peril in giving the public the arbitrating role too. The public is not a singular mind, after all, and so may hold conflicting understandings about the conferring of authority (or any other matter). Besides, those who constitute the public may themselves be interested parties, and so may have a stake in how the matter gets settled.

A more generous reading to Holmes is to take "the public" as a kind of quasi-mind, a viewpoint taken from the general good. If we idealize the observer to the conferring of authority, we can abstract away any contingent interest such an observer has in the proceedings at issue, and thus arrive at an unsullied understanding of the conferral of obligation.

But pragmatists tend to hate that kind of abstract move; more on that when I look at Holmes' "Path of the law" essay.

More stacker toys

More video posts (they run 11 total--I'll clean all that up later...) at The Smokr Tumblr.

Stacker toys

Stacker toys raise a number of logically interesting challenges, especially if you try to use them as logical models and in logic games. A first look is at The Smokr Tumblr (5 video posts).

CFP--Logical Aspects of Computational Linguistics

LACL 2014
IRIT, Toulouse, France
18-20 June 2014
http://www.irit.fr/LACL2014/

FIRST CALL FOR CONTRIBUTIONS

LACL'2014 is the 8th international conference on logical and formal methods in computational linguistics. This conference addresses the use of type theoretic, proof theoretic and model theoretic methods for describing natural language syntax and semantics, as well as the implementation of natural language processing software relying on such models. It will be held at IRIT, Toulouse, France, from 18 to 20 June 2014.

Scope: 

Computer scientists, linguists, mathematicians and philosophers are invited to present their work on the use of logical methods in computational linguistics and natural language processing, in natural language analysis, generation or acquisition.

Topics of either theoretical or applied interest include, but are not limited to:

* logical foundation of syntactic formalisms
          o categorial grammars
          o minimalist grammars
          o dependency grammars
          o tree adjoining grammars
          o model theoretic syntax
          o formal language theory for natural language processing
          o data-driven approaches
* logics for semantics of lexical items, sentences, discourse and dialog
          o discourse theories
          o Montague semantics
          o compositionality
          o dynamic logics
          o game semantics
          o situation semantics
          o generative lexicon
          o categorical semantics
* applications of these models to natural language processing
          o software for natural language analysis
          o software for acquiring linguistic resources
          o software for natural language generation
          o software for information extraction
          o inference tasks
          o evaluation
          o scalability

 
SUBMISSIONS and PROCEEDINGS

Articles should be written in the LaTeX format of LNCS by Springer (see authors instructions at http://www.springer.com/computer/lncs?SGWID=0-164-6-793341-0) and may consist of up to 12 pages of content (including figures, bibliography, possible appendices). In exceptional cases by prior agreement with program committee a paper up to 16 pages may be considered. It is expected that each accepted paper be presented at the meeting by one of its authors.

Submission is exclusively admitted electronically, in PDF format, through the EasyChair system. The submission site is https://www.easychair.org/conferences/?conf=lacl2014

PROCEEDINGS

Accepted papers will be published as a volume of the FoLLI sub-line of Lecture Notes in Computer Science (LNCS) by Springer (http://www.springer.com/lncs).

Invited speakers:

Prof. Zhaohui Luo,  Computer Science Department, Royal Holloway, University of London, U.K.

Prof. Michael Moortgat,  Department of Language, Literature and Communication, Utrecht University, the Netherlands

Prof. Reinhard Muskens, Department of Philosophy, Tilburg University, the Netherlands

PREVIOUS EDITIONS

A selection of the 1995 articles appeared in a special issue of the Journal of Logic, Language and Information (7:4, 1998). The proceedings of the international conferences LACL'96, LACL'97, LACL'98, LACL'2001, LACL'2005, LACL'2011 appeared in the series Lecture Notes in Artificial Intelligence (volumes 1328, 1582, 2014, 2099, 3492, 6736), and the proceedings of LACL 2012 in Lecture Notes in Computer Science (volume 7351) published by Springer. 

IMPORTANT DATES

Paper submission deadline: March 15, 2014
Notification of acceptance: April 5, 2014
Camera ready copies due: April 15, 2012
Conference dates: June 18-20, 2014

CONTACTS

soloviev (at) irit (dot) fr and asher (at) irit (dot) fr, co-chairs of LACL 2014