boundary.41 Debate about law in the mid-twentieth century took on an
ideological element, informed by experience with āimmoralā Nazi laws
and the Cold War. It was later inļ¬‚uenced by the intervention of federal
courts, led by the Supreme Court, to end state-sponsored racial segrega-
tion as well as to assert new rights of personal privacy and fair criminal
procedures under the authority of the Constitution. The status of āprin-
ciplesā ā“ the question of whether they lie inside or outside the deļ¬nition
of law ā“ was engaged with these developments.
For his skepticism about principles Holmes was posthumously criti-
cized as an amoral authoritarian positivist, obscuring his true position.42
The same theme has more recently been sounded by Alschuler and others
as a condemnation for promoting ālaw without values.ā43 His conception
throughout, consistent with classical common law theory, has been anti-
thetical to a strict boundary. He maintained that morals and values work
their way into the law, but this did not provide judicial license for sitting
judges to import their own subjective values under the abstract language
While sounding like a positivist separation of law and morals, this is
distinct from legal ontology. It is a critique of abstraction and a criterion
of interpretation. While analytical in character, it is grounded in both
analytical and historical research. Through a combination of analytical
and historical methods, Holmes came to deny the positivist assumption
that law can be understood through taxonomic analysis, as a static or
The early research papers are dense and often awkwardly written, but
they are necessary to a full understanding of Holmesā™s intellectual path.
It is not a direct linear progression, but jumps from topic to topic and
doubles back to earlier insights. Nor is it obvious that Holmes always had a
clear goal in mind during this period, beyond absorbing and abstracting
as much information as possible. His work during the formative period
41 Lon L. Fuller, The Morality of Law (New Haven: Yale University Press, 1969); Dworkin,
Taking Rights Seriously.
42 Fuller, The Law in Quest of Itself (Boston: Beacon Press, 1966), 62, 117.
43 Alschuler, Law without Values.
44 Holmes, āThe Path of the Law,ā in Collected Legal Papers, 170.
45 We should distinguish the legal positivism growing out of John Austinā™s command deļ¬-
nition from the original and much broader āpositivismā of Auguste Comte, whose cause
Holmes had supported from college days. It is undeniable from the early essays that
Holmes hardly viewed himself as a strict follower of John Austin.
Holmes, Legal Theory, and Judicial Restraint
gives the impression of a record of explorative self-education, writing out
his thoughts as the most effective means of achieving an understanding
of his subject as a whole.
Harvard Law School, which he entered in the fall of 1864, did not hold
his attention; he left in December 1865 for private practice in Boston,
commuting back and forth from an active intellectual life in Cambridge.
He soon took upon himself a major responsibility, revising and updating
the leading four-volume treatise, Kentā™s Commentaries on American Law.
Amidst this formidable task, which obliged him to spend every spare hour
reading cases and textbooks covering the entire ļ¬eld of law, he drafted a
series of essays published over the decade 1870ā“80 in the American Law
Review, of which he served as an editor from 1870 to 1874.46
In editing Kentā™s Commentaries while simultaneously reading and re-
reading Austinā™s extended treatise, Holmes was undertaking an account
of the actual common law of his time, as seen beneath the shadow of
the most inļ¬‚uential theoretical analysis of law since Hobbes. The essays
are uniļ¬ed by this comparison in two phases. Holmesā™s attention was
ļ¬rst engaged by topics that provided him with historical and analytical
background for testing Austinā™s basic assumptions and conclusions. It
then shifted to seeking a groundwork for an alternative theory of his
own, in which major positivist precepts were rejected. The resulting con-
clusions were embodied in The Common Law, but there was to be more
development in his thinking while on the highest appellate courts of
Massachusetts and the United States, especially in constitutional law.
46 White, Justice Oliver Wendell Holmes, 87ā“147.
Holmes and Legal Classiļ¬cation
[E]very positive law, or every law strictly so called, is a direct or circuitous
command of a monarch or sovereign number to a person or persons in
a state of subjection to its author. And being a command (and therefore
ļ¬‚owing from a determinate source), every positive law is a law proper, or a
law properly so called.
Besides the human laws which I style positive law, there are human laws
which I style positive morality, rules of positive morality, or positive moral
In consequence of the frequent coincidence of positive law and morality,
and of positive law and the law of God, the true nature and fountain of
positive law is often absurdly mistaken by writers upon jurisprudence. (162)
John Austin, The Province of Jurisprudence Determined (1832)
The decade of the 1870s, the ļ¬rst part of which Holmes took up the
subject of legal classiļ¬cation, demonstrated a seriousness of purpose that
seems incongruent with the blithe spirit who half a century later would
write to his friend Frederick Pollock:
[ Justice] Brandeis the other day drove a harpoon into my midriff with reference
to my summer occupations. He said you talk about improving your mind, you
only exercise it on the subjects with which you are familiar. Why donā™t you try
something new, study some domain of fact? Take up the textile industries in
Massachusetts and after reading the reports sufļ¬ciently you can go to Lawrence
[Mass.] and get a human notion of how it really is. I hate facts. I always say the chief
end of man is to form general propositions ā“ adding that no general proposition
is worth a damn. Of course a general proposition is simply a string for the facts
and I have little doubt that it would be good for my immortal soul to plunge into
Holmes, Legal Theory, and Judicial Restraint
them, good also for the performance of my duties, but I shrink from the bore ā“
or rather I hate to give up the chance to read this and that, that a gentleman
should have read before he dies. I donā™t remember that I ever read Machiavelliā™s
Prince ā“ and I think of the Day of Judgment.1
Here in private correspondence is an unserious and somewhat misleading
Holmes. Although he may have been honest about a bumpy motorcar trip
to Lawrence from his summer home in Beverly Farms to study factory con-
ditions, he could not have hated facts to accomplish what he had done in
the 1870s. The reference to general propositions alludes to those earlier
days. Holmes had discussed the relation of facts and abstractions, particu-
lars and generals, with Wright, Green, James, Joseph Warner, Peirce, and
other amateur philosophers of Cambridge, contemporaneously with his
examination of the relation of speciļ¬cs and generalizations in the law.2
His close friend āWillā James had written from Berlin in 1868, āWhen
I get home letā™s establish a philosophical society to have regular meet-
ings and discuss none but the very tallest and broadest questions ā“ to be
composed of nothing but the topmost cream of Boston manhood. It will
give each one a chance to air his own opinion in a grammatical form,
and to sneer and chuckle when he goes home at what damned fools all
the other members are ā“ and may grow into something very important
after a sufļ¬cient number of years.ā Evidence of meetings is fragmentary
and comes mainly from Peirceā™s later recollections.3 A letter from Henry
James to Charles Eliot Norton in early 1872 mentions that Holmes, along
with his brother James ā& various other long-headed youths have com-
bined to form a Metaphysical Club, where they wrangle grimly & stick to
the question. It gives me a headache merely to know of it.ā4 Had Holmes
continued to keep the daily record of appointments and discussions that
he kept only from 1866 to late December 1867 (noting the philosophical
āpalaversā with Wright and James), we might have a clearer picture.5
1 Holmes to Pollock, May 26 1919, Holmesā“Pollock Letters, vol. 2, p. 14.
2 Fisch, āMetaphysical Club,ā 20.
3 Id., 3ā“29.
4 Id., 4ā“5.
5 Although the original club may have died in late 1872, we know from correspondence that
it was revived in 1876; Francis Ellingwood Abbot received a letter from William James in
January of that year inviting him āto join a Club for reading and discussing philosophical
authors, which meets once a week at present and is composed of C. C. Everett, N. St. John
Green, O. W. Holmes, Jr., John Fiske, Thos. Davidson. J. B. Warner, Prof. Bowen, and one
or two others [Peirce was in Europe until August]. We have begun with Humeā™s Treatise
on Human Nature and the next meeting is at this house, next Sunday evening at 1 2 past
seven promptly.ā Holmesā™s copious diary of his reading records Humeā™s Treatise in April
Holmes and Legal Classiļ¬cation 63
The term āpragmatismā was not formally applied to the thoughts of
these contemporaries until 1898, and Holmes never heard it applied to
such discussions.6 Certainly addressed was the nature of generals and
their relation to the particulars for which they served (in Holmesā™s apt
phrase) as a āstring.ā Pragmatism is characterized as the view that theories
must be linked to experience or practice,7 and as such the validity of a
proposition is made dependent on both its past and its future, in its
success as a reļ¬‚ection of experience and a guide to practice.
The historical context tells us much. Religion and Continental philos-
ophy were formative inļ¬‚uences on the currents of thought in nineteenth-
century Cambridge, and thence on the emergence of pragmatism.8
A leading issue was the status of empirical science in religious belief.
Descartes, Locke, Berkeley, and Hume were familiar authorities. Unitari-
anism, a liberal belief that accommodated the more commercial and sec-
ular culture of post-Calvinist New England, had found a defense against
Transcendentalist and other challenges in Scottish realism, a āphilos-
ophy of common sense.ā9 After Kant, debates over empiricism became
increasingly troubled over the status of supra-empirical knowledge.10 The
Scot William Hamilton, who inļ¬‚uenced the young Chauncey Wright, had
expounded an empiricism acceptable in Harvard and Cambridge, one
that left room for God and scripture without too much extra space for
1876, presumably in ļ¬nishing it. See Little, āEarly Reading,ā 192; Fisch, āMetaphysical
When Charles Hartshorne began preparing Peirceā™s Collected Papers in 1927, he wrote
to Justice Holmes, at eighty-six the only surviving member of the original club, to inquire
about it and received this reply: āI am afraid that I cannot help you much in the way of
recollections of Charles Peirce. I think I remember his father saying to me, āCharles is a
genius,ā and I remember the august tone in which, at one of the few meetings at which I
was present, Charles prefaced his opinion with ā˜Other philosophers have thought.ā™ Once
in a fertilizing way he challenged some assumption that I made, but alas I forget what.
But in those days I was studying law and I soon dropped out of the band, although I
should have liked to rejoin it when it was too late. I think I learned more from Chauncey
Wright and St. John Green, as I saw Peirce very little.ā Fisch, āMetaphysical Club,ā 10ā“11.
6 See Sami Pihlstrom, āPeirceā™s Place in the Pragmatist Tradition,ā in Cheryl Misak, ed., The
Cambridge Companion to Peirce (Cambridge: Cambridge University Press, 2004), 27ā“48.
In 1908 Holmes called pragmatism an āamusing humbugā in a letter to Pollock. Holmes
to Pollock, June 17, 1908, Holmes-Pollock Letters, vol. 1, p. 139.
7 Cheryl Misak, āCharles Sanders Peirce (1839ā“1914),ā in Cambridge Companion to Peirce, 1.
8 See Kuklick, Rise of American Philosophy, 5ā“27.
9 Id., 10ā“16.
10 Id., 16.
11 Id., 16ā“18.
Holmes, Legal Theory, and Judicial Restraint
This accommodation was disrupted by two developments, noted in
chapter 3, not long before Holmes returned from the Civil War: John
Stuart Millā™s skeptical attack on Hamilton, and Darwinā™s Origin of Species.
While Asa Gray, Harvard professor of natural history, would claim with
contemporary fundamentalists, āTo us a fortuitous Cosmos is simply
inconceivable. The alternative is a designed Cosmos,ā no one could post-
pone the destruction of the philosophical basis of Unitarian orthodoxy,
with its brittle understanding of empiricism, extending to its account
of the status of and basis for scientiļ¬c knowledge, including āgeneral
propositions.ā12 Concern with this problem was a main current energiz-
ing Wright, Peirce, James, Green, Holmes, and other members of the
club that James had proposed in 1868.
There appear to be three main interests brought to the conversations:
Wrightā™s evolutionary theorizing, Greenā™s esteem for Alexander Bainā™s
deļ¬nition of belief as that on which one is prepared to act, and the
legal theories of Green and Holmes. Green urged the others to look to
āthe practical signiļ¬cance of every proposition,ā and claimed that āevery
form of words that means anything indicates some sensible fact on the
existence of which its truth depends.ā13
Another current of special concern to Holmes at this time was the
context of John Austinā™s revolutionary jurisprudence. After deciding on
a career in law rather than philosophy,14 the status of legal concepts ā“ gen-
eral propositions in the law ā“ occupied his attention. Austin had brought
a new approach to expository jurisprudence in England.15 Holmes saw
at once that Austinā™s search for a uniform universal classiļ¬cation, if suc-
cessful, might transform law into a distinctive kind of empirical science.
If unsuccessful, it might oblige a different hypothesis on the relation of
facts to general concepts.
His objections to Austin would emerge in the essays covering the period
between 1870 and the Lowell Lectures in 1880. They have much to do
with the attempt to establish a coherent logical arrangement for law that
would place a discrete boundary around it with the objective of a uni-
versal comprehensive analysis. These elements, the notion of boundary
and of internal analysis, were not as important for Hobbes, and they had
12 Id., 20ā“26.
13 Id., 49; Fisch, āMetaphysical Club,ā 20, 26.
14 Novick, Honorable Justice, 95ā“96.
15 Herget, American Jurisprudence, 12ā“22.
Holmes and Legal Classiļ¬cation 65
never been attempted with Austinā™s scope and detail. Hobbes had led
the way with his identiļ¬cation of law as sovereign commands, but with
Austin, legal positivism fully adopts the characteristic that it has main-
tained to the present: the ideal of law, and of a philosophy of law, as (in the
words of Ronald Dworkin) an āautonomous, analytic, and self-contained
The issue most allied with the debate over legal reasoning in the late
tracts of Hobbes and Hale is that of sovereignty. The importance of a clear
location and identity for sovereign power was plain to Hobbes, although
given the nature of parliamentary participation in the sixteenth century
he could not solve the problem cleanly. Much debate was devoted by
Hobbes and his critics to deļ¬ning the locus of supreme power to make
the laws.17 Hale in the Reļ¬‚ections takes issue with certain aspects of this
account reļ¬‚ected in Hobbesā™s late Dialogue, in particular with his lack
of attention to the historical consensual limitations attributable to the
unwritten English constitution.18 Similar issues would be taken up by
Holmes in response to Austinā™s deļ¬nition of the sovereign power two
centuries later. To Holmes, writing in 1870, the command deļ¬nition
simply solved one deļ¬nitional problem by creating another: āIn the ļ¬rst
place, who has the sovereign power, and whether such a power exists at
all, are questions of fact and of degree.ā19
The center of the debate had begun to shift away from the source of
raw power toward identiļ¬cation of the conceptual framework and bound-
ary of law, a movement that would develop considerably further by the
time of Fullerā™s challenge to Hart at mid-twentieth century, becoming
still more extended today. Indeed, the two matters, the source of power
and the boundaries of positive law, are alike in important respects. Both
appear as matters of empirical fact from which important conclusions
must arise. They offer an identiļ¬ble and authoritive character to law, as
well as the prospect of understanding what is most basic in it.20 Con-
ļ¬dence in such essentials would presumably promote the effectiveness
of the legal order itself, its beneļ¬cial operation and improvement. It
16 See Dworkin, Justice in Robes, 188.
17 Yale, āHobbes and Hale,ā 135ā“50.
18 Hale, Reļ¬‚ections, 507ā“12.
19 Holmes, āCodes, and the Arrangement of the Law,ā in Formative Essays, 80, and in 1
Collected Works, 214.
20 See Roger A. Shiner, Norm and Nature: The Movements of Legal Thought (Oxford: Clarendon
Press, 1992), 5ā“9.
Holmes, Legal Theory, and Judicial Restraint
should expose vulnerabilities and increase the prospect of fulļ¬lling the
lawā™s purposes, the promotion of a more coherent system of justice.
With Austin, sovereignty and boundary are still linked by the command
deļ¬nition, but there is an important new element: logical arrangement.
This had not seemed imperative to Hobbes in his defense of the sovereign
prerogative. But if there is a weakness in Hobbesā™s position it is to be found
in the remark made by the Philosopher in response to the Lawyerā™s insis-
tence on reason inherent in the common law: āThere is not amongst men
a universal reason agreed upon in any nation, besides the reason of him
that hath the sovereign power. Yet though his reason be but the reason
of one man, yet it is set up to supply the place of that universal reason.ā
Hobbesā™s sovereign reason drew on the image of a discrete monarch and
was vulnerable to the charge of arbitrariness. It demanded elucidation by
later theorists interested in the Hobbesean project, fortifying and ratio-
nalizing the legal authority of the state. Addressing this deļ¬ciency in the
seventeenth century was the reformer Jeremy Bentham, with the theory
Austin was a loyal disciple of Bentham, and there is a lengthy treat-
ment of utilitarianism in his Lectures.22 There is also a direct connection
between that discussion and Austinā™s choice of logical arrangement, based
on various types of right and duty found throughout the law.23 While rec-
ognizing the close relationship of duties and rights, Austin apparently
preferred the latter out of concern to emphasize the positive scope of
action necessary to maximize utility:
But the ļ¬nal cause or purpose for which government ought to exist, is the fur-
therance of the common weal to the greatest possible extent. And it must mainly
attain the purpose for which it ought to exist, by two sets of means: ļ¬rst, by con-
ferring such rights on its subjects as general utility commends, and by imposing
such relative duties (or duties corresponding to the rights) as are necessary to the
enjoyment of the former: second, by imposing such absolute duties (or by imposing
21 As Friedrich observes, Hobbes sought to mitigate the arbitrariness of sovereign power
by assigning judges the task of interpreting a ālaw of nature,ā and his legal philosophy is
based on a principle of utility. Friedrich, Philosophy of Law in Historical Perspective, 87, 89.
22 Austin, Lectures, vol. 1, pp. 109ā“70.
23 Id., 412: āA monarch or sovereign body expressly or tacitly commands, that one or more
of its subjects shall do or forbear from acts, towards, or in respect of, a distinct and
determinate party. The person or persons who are to do or forbear from these acts, are
said to be subject to a duty, or to lie under a duty. The party towards whom those acts
are said to be done or forborne is said to have a right, or to be invested with a rightā
(emphasis in original).
Holmes and Legal Classiļ¬cation 67
such duties without corresponding rights) as tend to promote the good of the
political community at large, although they promote not specially the interest of
Holmes too was drawn to the prospect of logical arrangement, for reasons
that would later appear ambivalent. Benthamā™s earlier call for codiļ¬ca-
tion, as a precondition for the reform of society and legal institutions, was
still much discussed in England and America, and codiļ¬cation was a live
issue in several states after the Civil War.25 Holmesā™s mistrust of the cod-
iļ¬cation project is evident in remarks made in the 1870 article, āCodes,
and the Arrangement of the Law,ā opening as it does with the reference
to Lord Mansļ¬eld and the case-speciļ¬c reasoning of the common law.
It is followed by the remark, āThese [aspects of common law method]
are advantages the want of which cannot be supplied by any faculty of
generalization, however brilliant, and it is noticeable that those books on
which an ideal code might best be modeled avowedly when possible lay
down the law in the very words of the court.ā26
Holmes appears from the outset suspicious of the Hobbesean or Aus-
tinian position in the debate over sovereign reason. Nevertheless, he
proceeded in 1870 to experiment with his own logical arrangement, pre-
ferring a classiļ¬cation under the rubric of duty rather than right, as duty
was empirically closer to the proscriptive nature of the law:
Duties precede rights logically and chronologically. Even those laws which in form
create a right directly, in fact either tacitly impose a duty on the rest of the world,
as, in the case of patents, to abstain from selling the patented article, or confer an
immunity from a duty previously or generally imposed, like taxation. The logical
priority of duty in such instances is clear when we consider that in its absence
any man might make and sell what he pleased and abstain from paying for ever,
without assistance from law. Another illustration is, that, while there are in some
24 Lectures, vol. 1, p. 282; see also 290ā“91: āEvery legal right is the creature of a positive law:
and it answers to a relative duty imposed by that positive law, and incumbent on a person
or persons other than the person or persons in whom the right resides. To every legal
right, there are therefore three parties: The sovereign government of one or a number
which sets the positive law, and which through the positive law confers the legal right,
and imposes the relative duty: the person or persons on whom the right is conferred:
the person or persons on whom the duty is imposed, or to whom the positive law is set
25 Horwitz, Transformation of American Law 1870ā“1960, 117, 123.
26 Holmes, āCodes, and the Arrangement of the Law,ā in Formative Essays, 77, and in 1
Collected Works, 212.
Holmes, Legal Theory, and Judicial Restraint
cases legal duties without corresponding rights, we never see a legal right without
either a corresponding duty or a compulsion stronger than duty.27
His criticism of the ārightsā system was that it was not close enough to
the actual operation of law and legal sanctions. It presented an image
of an envelope of protection, emphasizing the scope of conduct made
permissible by law, whereas that of duties was of conduct merely prohi-
bited or required. Holmes concluded that the latter was closer to the true
picture. This is nearer the vision of Coke and Hale than of Hobbes and
Austin; the scope of possible social conduct had no necessary relation
to the phenomenon called law. It could be engaged in anyway, and was
neither created by nor dependent on state sanction, nor any hypothetical
legitimating concept, such as rights.
While Holmes was writing this critique, Green was boiling down cer-
tain legal concepts to their practical essence. In āProximate and Remote
Cause,ā also published in 1870, he concluded that liability for personal