<< . .

( 23)

. . >>

injury was not the physical remoteness from the defendant™s conduct but
the degree of certainty with which injury might have been anticipated.
After Green™s early death in 1876, a year after Wright™s, Holmes would
take this insight to its fullest extension.28
In his earlier writings of the decade 1870“80, Holmes experimented
with what he called the “philosophical” organization of the different
branches of law seen as classi¬cations of duty (see the Appendix). In
the course of so doing, his initial doubts concerning sovereign command
and sanction gradually matured in a manner that was to transform his
ultimate perspective into an evolutionary one. By 1872, using a review of
an article by Frederick Pollock to summarize a series of lectures he had
given to undergraduates at Harvard College, Holmes reached a more
explicit critique of Austin™s command de¬nition, speci¬cally in response
to Austin™s rejecting custom as part of law. As pure custom was neither
commanded nor sanctioned by the sovereign, Austin considered it but a
“motive for decision,” becoming law only when its adoption by the legis-
lature or the courts demonstrated the tacit consent of the sovereign. To
this Holmes replied:
Austin said, following Heineccius (Recitationes, §72), that custom only became
law by the tacit consent of the sovereign manifested by its adoption by the courts;
and that before its adoption it was only a motive for decision, as a doctrine of

27 Holmes, “Codes, and the Arrangement of the Law,” in Formative Essays, 79“80; 1 Collected
Works, 214.
28 Kuklick, Rise of American Philosophy, 49“50.
Holmes and Legal Classi¬cation 69

political economy, or the political aspirations of the judge, or his gout, or the
blandishments of the emperor™s wife might have been. But it is clear that in many
cases custom and mercantile usage have had as much compulsory power as law
could have, in spite of prohibitory statutes; and as to their being only motives
for decision until adopted, what more is the decision which adopts them as to
any future decision? What more indeed is a statute; and in what other sense law,
than that we believe that the motive which we think that it offers to the judges
will prevail, and will induce them to decide a certain case in a certain way, and so
shape our conduct on that anticipation?29

While this passage has frequently been interpreted as an early statement
of the attitude later characterized as legal realism, it is noteworthy for the
comment that “custom and mercantile usage have had as much compul-
sory power as law could have, in spite of prohibitory statutes.” This too is
an extension of the approach of Hale and Blackstone, quite at odds with
that of Hobbes and Austin. Custom is already at work with a “compulsory
power.” Law is not objecti¬ed as separate from the factors motivating
decision, and custom is indeed placed on the same footing as statutory
law, even as Blackstone had equated the lex scripta with the lex non scripta.
Austin, on the other hand, had drawn a sharp distinction between cus-
tom and statutes, holding that unsanctioned custom was merely a rule of

Now a merely moral, or merely customary rule, may take the quality of a legal rule
in two ways: “ it may be adopted by a sovereign or subordinate legislature, and
turned into a law in the direct mode; or it may be taken as the ground of a judicial
decision, which afterwards obtains as a precedent; and in this case it is converted
into a law after the judicial fashion. In whichever of these ways it becomes a legal
rule, the law into which it is turned emanates from the sovereign or subordinate
legislature or judge, who transmutes the moral or imperfect rule into a legal or
perfect one.30

Holmes then takes up the question that, since Hobbes, has connected
sovereignty with the boundary of law: the de¬nition of law as the com-
mand of the sovereign:

Passing to the suf¬ciency of Austin™s de¬nition for determining what sovereign
commands are to be called law . . . the speci¬c penalty or sanction which Austin
seemed to tacitly assume as the ¬nal test, could not always be relied on.
The notion of duty involves something more than a tax on a certain course of
conduct. A protective tariff on iron does not create a duty not to bring it into the
country. The word imports the existence of an absolute wish on the part of the

29 Holmes, “Book Notice,” in Formative Essays, 91“92, and in 1 Collected Works, 294, 295.
30 Austin, Lectures, vol. 2, p. 553.
Holmes, Legal Theory, and Judicial Restraint

power imposing it to bring about a certain course of conduct, and to prevent the
contrary. A legal duty cannot be said to exist if the law intends to allow the person
supposed to be subject to it an option at a certain price. The test of a legal duty
is the absolute nature of the command.31

Here the criticism by Hale, of the dubious location by Hobbes of the
sovereign power and its problematic identi¬cation, is given a still further
elaboration. Addressing the overall character of the law itself, Holmes
argued that various manifestations are not of the nature of commands.
It is not just the dif¬culty of establishing a clear identi¬cation of the
sovereign that undermines positivist analysis; Holmes now points to
aspects of law that are empirically antithetical to the analytical project.
While Holmes™s emphasis on the concept of duty re¬‚ects the central-
ity Austin gave to both duties and rights in his systematic account of
jurisprudence, he had begun to challenge the project of any such logi-
cal arrangement. H. L. A. Hart would make a criticism similar to that of
Holmes nearly a century later: “Legal rules de¬ning the ways in which
valid contracts or wills or marriages are made do not require persons to
act in certain ways whether they wish to or not. Such laws do not impose
duties or obligations.”32 But whereas Hart would restructure legal posi-
tivism by placing the concept of law within a new boundary consisting of
primary and secondary rules and a Rule of Recognition,33 Holmes would
proceed in a direction that would lead him to abandon the quest for
an analytical system for all law and to question whether any determinate
boundary could be established at all.
Before addressing this, we should note the occasion for the short 1872
article in which Holmes reviewed an article by Frederick Pollock. The
title of Pollock™s piece was “Law and Command,” and it was an extended
criticism of the command de¬nition:

It will be my aim to show that this de¬nition, if exclusively insisted on, errs by ele-
vating what is at most one characteristic of law into its essence, that contrariwise,
by losing sight of what is really an essential constituent, it narrows the proper
scope of law and tends to an unsatisfactory view of its operation, and that by
putting forward the arbitrary and suppressing the necessary aspect of legislation
it seriously obscures the organic relation of law to the community.34

31 Holmes, “Book Notice,” in Formative Essays, 92, and in 1 Collected Works, 296.
32 Hart, Concept of Law, 27.
33 Id., 77“96.
34 Pollock, “Law and Command,” Law Magazine and Review, new series, no. 3 (1872), 191.
Holmes and Legal Classi¬cation 71

Arguing against a strictly analytical approach, Pollock emphasized the
continuing in¬‚uence of custom on the law “ the same point that custom
is already at work independent of sovereign command with its own com-
pulsory power. Pollock urged an alternative theory recognizing that law
evolves, a point that Holmes would emphasize in the later phase of his
own research:

Law in the widest sense is a condition, or assemblage of conditions, under which
the evolution of things proceeds, . . . and the determination of which is part of
the collective consciousness of that society.35

In 1873 Holmes published “The Theory of Torts,” in which the com-
mon law model was enhanced with the previously quoted (in chapter 3)
depiction of decisions accumulating around opposing poles. Although
apparently moving away from the notion of arranging or schematizing all
law around the concept of duty, and toward a theory of liability more like
that of Green™s, he would again publish a revised “expository” table of
duties (see the Appendix). The connection between the early and later
articles has not been altogether clear.36 If the direction of the early arti-
cles is dif¬cult to fathom now, it presented dif¬culty when ¬rst written to
no less a scholar than Frederick Pollock.
Holmes had been introduced to Pollock during a visit to England
in the summer of 1874, and the two became lifelong friends. There is
much evidence in their ¬fty-eight-year correspondence of common views
about law. Yet Pollock in 1877 asked whether Holmes thought a codi¬ed
arrangement was undesirable “in itself ” (Pollock™s emphasis) or only “that
there is no advantage in doing it by legislative authority.” He confessed,
“I am not really in possession of your view” and suggested it be made
plainer in a future article.37 Holmes™s reply, if any, has not survived, but his
published writing had already begun to move beyond this issue. By 1876
he had satis¬ed himself that a comprehensive philosophical arrangement
was impossible and had shifted his attention from right and duty to the
evolving nature of legal concepts in general. Though he had already sent

35 Id., 205.
36 See Herget, American Jurisprudence, 41: “But Holmes™ ¬‚irtation with the expository tradi-
tion was short-lived. He became convinced in the latter part of the 1870s that the pursuit
of jurisprudence in this direction would lead to deadends, or worse. A perceptive biog-
rapher of Holmes, Mark DeWolfe Howe, suggests that Holmes perceived the mystical
hand of German metaphysics behind the expository paradigm, and this ran counter to
his empiricist and pragmatic leanings.”
37 Pollock to Holmes, July 26, 1877, Holmes-Pollock Letters, vol. 1, pp. 7“8.
Holmes, Legal Theory, and Judicial Restraint

Pollock all the articles published through 1876, Pollock failed to see the
shift, which is found in a close critique of Austin.
Austin had in his Lectures on Jurisprudence stressed the relation between
the critical concepts of rights and duties to the status of persons affected
by them:

There are certain rights and duties, with certain capacities and incapacities to take
rights and incur duties, by which persons, as subjects of law, are variously deter-
mined to certain classes.
The rights, duties, capacities, or incapacities, which determine a given person
to any of these classes, constitute a condition or status which the person occupies,
or with which the person is invested.
One and the same person may belong to many of the classes, or may occupy, or
be invested with, many conditions or status. For example, one and the same per-
son, at one and the same time, may be son, husband, father, guardian, advocate,
or trader, member of a sovereign number, and minister of that sovereign body.
And various status, or various conditions, may thus meet or unite in one and the
same person, in in¬nitely different ways. (emphasis in original)38

To understand his focus, it is important to remember that Austin had in
mind the creation of a table that would effectively and comprehensively
display the arrangement of the essential classi¬cations of the law. Austin™s
published lectures contained various tables, but none are dispositive in
detail or vindicate his project.39 Holmes, in an article published in 1872
that still experiments with the duty scheme, “The Arrangement of the
Law: Privity,” set forth his own table (see the Appendix) suggesting that
the task is impossible. His claim is that, because rights and duties “may
be succeeded to by another who cannot ¬ll the situation” (on which
the original status was based), the task of organizing the law around the
concepts of right and duty is impossible. “It is obvious,” notes Holmes of
his own table, “that this scheme does not exhaust the whole body of the
Holmes™s second attempt to arrange the law under categories of duty in
1872 encountered a distinct threat to the prospective comprehensiveness
of the overall project of logical arrangement. To see whether his system
of duties could be applied across the board, Holmes, yet focusing on
the methodological implications of Austin™s Lectures, developed a chart

38 Austin, Lectures, vol. 2, p. 706.
39 Austin, Lectures, vol. 1, p. 79.
40 Holmes, “The Arrangement of the Law: Privity,” in Formative Essays, 96, 97, and in 1
Collected Works, 304, 305; see Appendix.
Holmes and Legal Classi¬cation 73

dividing the law by reference to the classes of persons on whom burdens
were imposed as well as to those in favor of whom they are imposed.
Holmes™s six divisions were “duties of all the world to the sovereign,” “to
all the world,” and “to persons in particular situations,” and also “duties
of persons in particular situations to the sovereign,” “to all the world,”
and “to other persons in particular situations.” When he proceeded to
test the arrangement, it became apparent that the conceptual scheme
ascribed a primary importance to what Holmes called “the situation of
fact” or the “de¬nition of the situation.”41

The duties to persons in a particular situation begin with their beginning, and
end with their ceasing, to ¬ll that situation. When you describe the situation, that
is, the facts, to which the duties are incident as a legal consequence, you describe
the beginning and end of the duties as to a given individual.42

A problem arose in considering legal succession. If classifying duties
depended on the situation of fact creating such duties, then succession
of others to those duties should in theory depend on succession to the
situation of fact. But while this might be true in the majority of actual
instances, it was by no means true of all:

Some continuing rights are incidents to a situation of fact, which can only be
¬lled by the ¬rst person entitled to the rights in question. A certain individual
and no other is the person with whom a certain contract was made, or to whom
a certain franchise or monopoly was granted; yet the continuing rights incident
to the situation of contractee or grantee may be succeeded to by another who
cannot ¬ll the situation, and the same is true of ownership as distinguished from
bare possession.43

This led Holmes to ask how the law had made possible succession by
others not party to the original situation and the ascription of the original
duties to or for the successor. Another way of putting the question was to
ask how the law had been able to continue using the absolute termino-
logy of duties, which implied a relationship between the individual and
the situation, in instances where the original de¬ning relationship did
not exist. His answer was that it had succeeded in doing so through the

41 Formative Essays, 114, and 1 Collected Works, 316. See Appendix. These concerns appar-
ently originate with Austin™s lectures on the relation of rights to status; see Lectures,
vol. 2, pp. 718“59.
42 Holmes, “The Arrangement of the Law: Privity,” in Formative Essays, 98, and in 1 Collected
Works, 304.
43 Formative Essays, 98, and 1 Collected Works, 304“6.
Holmes, Legal Theory, and Judicial Restraint

creation, at an earlier time, of a ¬ctitious identi¬cation of the successor
with the original person.

The aggregate of the ancestor™s rights and duties or total persona sustained by
him was easily separated from his natural personality, and regarded as sustained
in turn by his heir, in view of the fact that it was originally his only as head of the
family, and consisted of the aggregate of the family rights and duties. If we start
here with succession to the entire situation of an individual in the community,
on the assumption of his entire persona, we shall ¬nd the other and more usual
examples of succession in privity easier to understand. . . . The ¬rst succession
in privity was the universal succession of the Roman law; then privity in the suc-
cession to speci¬c things occurs when the notion of ownership was originally
subordinate to a personal relation with the right over a thing as an incident, then
it is extended to successions generally.44

In the course of researching this issue, Holmes found the history of the
relation between master and servant to be particularly illuminating of the
phenomenon, although constituting a special case:

We have thus far dealt with clear cases of substitution where a successor assumes
a persona to the exclusion of the individual who had sustained it until then. There
is another class, where the new comer is introduced under a persona without
excluding his predecessor.45

While a servant eventually assumed a legal status independent of his
master, this was not originally true:

Under the early Roman law the wife, children, and servants of a citizen were his
slaves. They could not be said to stand in a legal relation to him, for they had no
standing before the law except as sustaining the persona of the family head.46

It was here that Holmes observed that the ¬ctitious identi¬cation of ser-
vant and master might also elucidate the origin of the doctrine of vicarious

It will be observed, moreover, that as the master™s right to bene¬ts acquired by
his servants is general, and as he is liable for the latter™s torts wherever a liability
is imposed, the slave may be said to sustain his master™s persona for purposes
inde¬nite not only in number but in kind.47

44 Formative Essays, 100, and 1 Collected Works, 307. The early family had been identi¬ed with
its head, and the heir™s assumption of the family headship with its rights and obligations
led to the ¬ction that a grantee assumed the grantor™s identity in the same manner,
eventually spreading to the law of chattels and other rights and obligations.
45 Formative Essays, 110, and 1 Collected Works, 313.
46 Formative Essays, 110, and 1 Collected Works, 314.
47 Formative Essays, 111, and 1 Collected Works, 314.
Holmes and Legal Classi¬cation 75

This ampli¬cation of the essay on privity was to become the connecting
link to Holmes™s next essay, “The Theory of Torts,” published in July 1873.
While in form it sought to set forth a new arrangement of duties implicit in
the branches of tort law consistent with the overall duty scheme, the essay
devotes much of its attention to bringing together the strands of earlier
doubt and weaving the outline of a new theme “ the growth of case law
through the gradual articulation of standards of conduct. Through the
master-servant example Holmes ¬rst con¬rmed the connection between
his early demonstration of the lack of coextensivity of duty and sanction
with its corollary that liability to a civil action does not import culpability.
The continuity with the previous essay is revealed in Holmes™s reference
to it:
I do not owe my butcher a duty not to buy his meat, because I must pay for it if
I buy. And as liability to a civil action for the amount of the plaintiff™s detriment
is quite different from a punishment proportioned to the defendant™s guilt, so,
conversely, liability to such an action does not necessarily import culpability, as it
has been thought to do by some of Bentham™s followers. The liability of a master
for his servant, which is one of the instances illustrative of this proposition, and
which Austin tried to account for by the notion of remote inadvertence, has been
explained heretofore.48

On securing this connection, Holmes set the stage for one ¬nal ap-
proach to classi¬cation based on duty, now by dividing torts into cate-
gories of duties in which the consciousness of the party liable is an element
and those in which it is not. In the latter category Holmes distinguished
between duties determined by acts or events exactly de¬ned and those
not exactly de¬ned. Into the latter category he placed “negligence latiori
sensu.” The majority of the essay is devoted to explaining how negligence
came to be put there, and it marks a turning point.
Negligence latiori sensu meant negligence in the broadest sense, cov-
ering all the cases in which it might be pleaded, not necessarily involving
proof of a particular state of mind. In dividing torts into duties wherein
consciousness was and was not an element, “negligence” cases in the
broad sense at ¬rst posed a dilemma. It had become clear from the
master-servant example that there were at least some within this group “
master-servant cases having been treated as negligence by lawyers as well
as by the Austinian school of jurisprudence “ that did not involve any
proof of the defendant™s lack of care. What then was to be done with the
entire group?

48 Holmes, “A Theory of Torts,” in Formative Essays, 117, and in 1 Collected Works, 326.
Holmes, Legal Theory, and Judicial Restraint

Half-way between the two groups which have been indicated [requiring and not
requiring consciousness] lie the great mass of cases in which negligence has
become an essential averment, since Bentham™s ideas have gotten into the air,
and the abolition of the old forms of action has allowed pleaders to state their
case according to their own views of its essential elements. What does this mod-
ern negligence mean? Austin, following his general notion that liability imports
culpability, analyzed negligence as the state of the defendant™s mind. This seems
to us unsatisfactory; and to show why, we must begin at a little distance from the

Presenting the question in this manner led Holmes to examine the entire
group from the standpoint of the development of legal precedents in neg-
ligence cases, and it brought him to the position that the entire group
should be placed in the latter category, those not requiring conscious-
ness of the defendant. Looking behind the practice of lawyers, and the
assumptions of analytical jurisprudence that were based on it, the law had
to be seen as a process of constant change:
The growth of the law is very apt to take place in this way: Two widely different
cases suggest a general distinction, which is a clear one when stated broadly. But as
new cases cluster around the opposite poles, and begin to approach each other,
the distinction becomes more dif¬cult to trace; the determinations are made
one way or the other on a very slight preponderance of feeling, rather than on
articulate reason; and at last a mathematical line is arrived at by the contact of
contrary decisions, which is so far arbitrary that it might equally well have been
drawn a little further to the one side or the other.50

The critical observation was that the submission of a case to the jury for
trial on the issue of negligence was, as a practical matter, simply part of the
process of the evolution of explicit standards of conduct in areas where
they had not been settled on by either statute or the growth of precedent.
Hereupon Holmes cited the Beadel v. Perry decision, for the proposition
of making explicit a rule emerging from prior cases by holding that “a
building cannot be complained of unless its height exceeds the distance
of its base from the base of the ancient windows.”
Thus he concluded that negligence was not a hybrid class between
conscious and unconscious tort but was governed generally by external
standards of conduct. Having reached this generalization, Holmes was
able to avoid an ambiguous place for negligence. This led to his threefold
division of tort law into the categories of “duties of all to all,” “persons in

49 Formative Essays, 118, and 1 Collected Works, 327.
50 Formative Essays, 119, and 1 Collected Works, 327.
Holmes and Legal Classi¬cation 77

particular situations to all,” and conversely of “all to persons in particular
situations.” And, one ¬nal time, Holmes was to allude to the original
focus that had launched his research:

Indeed it is believed to be one of the evils of not having a comprehensive arrange-
ment of the law that we lose the bene¬t of such generalizations as a philosophical
system would naturally suggest, and cases are discussed only on the foot of the
particular relation out of which they arise dramatically, but of which they are

<< . .

( 23)

. . >>

Copyright Design by: Sunlight webdesign