Encyclopedia of the Philosophy of Law and Social Philosophy

Living Edition
| Editors: Mortimer Sellers, Stephan Kirste

American Realism - Development and Critique

  • Matthew X. EtchemendyEmail author
Living reference work entry

Latest version View entry history

DOI: https://doi.org/10.1007/978-94-007-6730-0_336-2


The label “American Legal Realism” stands not for a discrete doctrine or single idea but for a movement in American legal thought that flourished in the interwar years and for the general “approach to law, adjudication, and legal education” associated with it (Fisher et al. 1993, xi). American Legal Realist thought has many aspects, but among the most distinctive and influential of these is the Realist perspective on the study of law and in particular the question of what should (and what should not) be the principal object and methods of a science of law. The present entry deals with this aspect of American Legal Realism.

The American Legal Realists of the 1920s and 1930s were a diverse group with a range of scholarly outlooks and interests (Kronman 1993, 185–186). In fact, there is not even a canonical, exhaustive list of Realists. (The figures cited herein as American Realists – Felix Cohen, Walter Wheeler Cook, Jerome Frank, Leon Green, Joseph Hutcheson, Karl Llewellyn, Underhill Moore, Herman Oliphant, Max Radin, and Hessel Yntema – are, however, all commonly regarded as part of the group.) A degree of generalization and philosophical reconstruction is therefore necessary if one is to speak broadly about the American Realist stance on any particular issue, including legal science. Be that as it may, it remains possible to trace through the writings of the major Realists (1) a distinctive critique of the dominant pre-Realist conception of legal science as well as (2) a distinctive, albeit in many ways skeletal, proposal for what to replace it with.

The Pre-realist Orthodoxy

The easiest way to explain the Realist stance on legal science is to contrast it with a certain once-dominant (and still persistent) way of thinking about the law and its study – a pre-Realist “orthodoxy” to which American Realism stands as a reaction or rebellion (Grey 1996, 495; Kronman 1993, 186; Leiter 2007, 87). And just what was the orthodoxy against which the Realists rebelled? As Frederick Schauer notes, “much writing in the Realist tradition…has aimed at caricatured and typically nonspecified targets” (Schauer 2013, 753n17), unnamed adherents of a so-called conceptualist or mechanical jurisprudence. But among the most common specific intellectual foils for the Realists were Christopher Columbus Langdell, the Dean of Harvard Law School from 1870 to 1895, and Joseph Henry Beale, “the most self-consciously philosophical exponent of classical orthodoxy” (Grey 1983, 29), also of Harvard Law (Schauer 2013, 753n17; Sebok 1995, 2071, 2077–2078). Langdell, in particular, is an important and useful figure with whom to contrast the Realists, because he was in many ways the father of modern American legal education (Kimball 2004, 277, 2009), and his conception of legal science offers a prominent, concrete example of the outlook the Realists sought to undermine.

Boiled down to its barest essentials, Langdell’s basic view on the study of law was as follows. The study of law is a science, and law, “considered as a science, consists of certain principles or doctrines.” The mark of a “true lawyer” is to master the principles and doctrines of the law and “to be able to apply them with constant facility and certainty to the ever-tangled skein of human affairs” (Langdell 1879, viii). The relevant principles and doctrines of the law have developed slowly over time, and the best way to learn them is to study reports of decided cases. Thus, the library is to legal scientists what the laboratory is to chemists or physicists (Langdell 1887, 124). The Langdellian legal scientist examines the reports of decided cases, abstracts principles and doctrines therefrom, and finally infers how these principles and doctrines are to be applied in new factual scenarios (Cook 1927, 307–308; Wambaugh 1906, 2). Apart from the vaguely grandiose analogy to the natural sciences, this conception of legal study should strike most contemporary graduates of American law schools as relatively familiar, especially as it pertains to traditional common-law subjects like contracts, property, and torts.

At its heart, the Langdellian science of law involves the mastery of a system of rules. The relevant rules are not descriptive generalizations (such as the “rule” that what goes up must come down) but standards against which actions or decisions can be compared or evaluated. For example: one venerable doctrine of contract law is that there is, in general, no contract without consideration (Langdell 1880, 58). The fact that this rule is part of the law does not mean, however, that judges will always decide in accordance with it, for judges sometimes fail to follow the law. Depending on one’s background philosophical commitments, this may sound a bit mystifying: can we take seriously a science whose object is a body of prescriptive rules, existing in an “extra-experiential” domain of abstract entities (Frank 1930, 54)? Such concerns may have merit, but the metaphysical commitments of Langdellian legal science are not necessarily extravagant by everyday standards. An abstract body of legal rules is not especially more metaphysically queer than, say, abstract bodies of rules governing chess or English grammar, whose existence most of us are inclined to accept. To be sure, the rules of chess and English grammar are, in some manner or another, dependent on convention or social fact, and they evolve over time. Still, we readily allow that there are such rules and that one can, through systematic study, become more proficient in stating and/or applying them.

In many ways, the rules and principles posited by the Langdellian orthodoxy can be understood along these lines (Grey 1983, 27n98). Langdell did not think of the law as static (Langdell 1879, viii) and saw it as rooted in social practice rather than, say, natural law (Sebok 1995, 2082). (The same can also be said of Beale (1916, 143, 149–150).). Common-law rules, though “unwritten” in the sense that they are not authoritatively promulgated once and for all, can be learned by studying legal practice, mainly via the reports of decided cases. While any one data point consulted by the legal scientist can be misleading (some cases are incorrectly decided and/or poorly reasoned, just like some English speakers utter ungrammatical sentences and some chess players break the rules), this is not in principle a bar to successfully abstracting the rules from the overall body of evidence. And the rules are, when properly understood, sufficiently determinate that a learned lawyer can apply them “with constant felicity and certainty” to new scenarios.

The Realist Critique

There are many legitimate sources of philosophical puzzlement about the Langdellian picture outlined above – for example, worries rooted in the metaphysics of rules. But rarefied philosophical considerations were not the primary factor the Realists emphasized in attacking the Langdellian model of legal science. Instead, the Realist critique largely hinged on the following, thoroughly pragmatic question: can the orthodox style of doctrinal study advanced by the likes of Langdell yield the knowledge needed for the successful practice of law (Leiter 2007, 90)? In order to answer this question, of course, one needs some conception of what the successful practice of law involves. On this point, the Realists were pretty clear. Like their spiritual father Oliver Wendell Holmes, Jr., the Realists emphasized that lawyers are hired to achieve practical results, not for their ability to expound an abstract body of legal rules per se (Frank 1930, 55; Holmes 1897, 457–462; Llewellyn 1930b, 446n12). As a general matter, clients want to know what will or will not provoke action on the part of courts, so they can plan their business; and if litigation commences, clients want lawyers to persuade judges to decide in their favor. (In strictness, this judiciary-centric formulation should be extended to include other institutional actors like administrative agencies (Llewellyn 1930b, 455–456), but it is simpler to speak mainly in terms of courts, as the Realists themselves tended to do.) As the Realists often observed, therefore, what is really crucial for lawyers is a good understanding of how courts will act under different circumstances (Cook 1927, 308; Frank 1932, 646; Llewellyn 1930a, 2–5; Moore and Hope 1929, 703; Radin 1925, 362), including in response to different arguments and litigation strategies. Knowledge of “legal rules and principles” per se is accordingly beside the point, except (and it is an important exception!) insofar as such knowledge can help lawyers predict and, derivatively, influence official action (Llewellyn 1930a, 4).

These observations do not, however, alone reveal any deep inadequacy in an orthodox Langdellian approach to the study of law. Suppose that (1) judges uniformly applied the law in a “rational, honest, competent, and error-free” manner (Leiter 2007, 9) and (2) the legal rules provided a determinate answer for what judges are to do in all relevant circumstances. As a shorthand, we can call these two suppositions “Condition 1” and “Condition 2,” respectively. Note that if both Conditions obtained, a lawyer with perfect knowledge of the legal rules could presumably predict judicial decisions quite ably. Of course, nobody has ever believed that both Conditions (especially Condition 1) hold across the board: that is plainly an idealization. But what if reality did not depart all that much from the Conditions? If so, there would be no obvious reason to be dissatisfied with the basic Langdellian model of legal science, assuming anyway that the legal rules can reliably be learned by studying the relevant sources, notably reports of decided cases.

A sufficiently systematic failure of either Condition 1 or Condition 2 would, however, justify serious doubts about the adequacy of the Langdellian approach and indeed any approach that placed exclusive focus on the mastery of legal rules or doctrine. Supposing a sufficient degree of irrationality, dishonesty, incompetence, and/or proneness to error on the part of the judiciary – that is, failure of Condition 1 – knowledge of the legal rules would not help much in predicting judicial action, even if the rules provided determinate answers to all cases. There simply would not be much reason to expect judges to decide in accordance with the rules at all. And supposing sufficient indeterminacy in the rules – that is, failure of Condition 2 – it would be impossible to predict judges’ actions reliably on the basis of those rules, even if judges uniformly applied them with perfect fidelity. For the rules would often provide no determinate answer, in which case judges would have to reach a decision on some other basis, whether or not they realized or admitted it.

So one could justify dissatisfaction with the Langdellian model of legal science by emphasizing failures of Condition 1, Condition 2, or both. Failure of Condition 1 alone would, however, justify a considerably less radical critique of Langdellian legal science than would failure of Condition 2 (Leiter 2007, 9–10). To be sure, if judges were sufficiently unreliable appliers of the legal rules, Langdellian training would not be very good preparation for the realities of legal practice. But a defender of Langdellian legal science could simply acknowledge the point and urge that we should appoint more honest and/or legally knowledgeable judges. Failure of Condition 2, however, would cut deeper. To the extent legal rules are indeterminate, judges cannot resolve cases solely by applying those rules, regardless of honesty or skill. Remedying the situation would then require, at minimum, altering the legal rules (perhaps radically) in order to realize a sufficient degree of doctrinal determinacy – a goal that, even if theoretically attainable, might well not be worth the cost (Frank 1930, 5–10, 138, 186–193).

Unsurprisingly, then, the Realists gave greater focus to the failure of Condition 2: although they did sometimes speak to failures of Condition 1, such as outright judicial corruption (Frank 1931, 33–35), the more cutting problem of legal indeterminacy was at the heart of their critique of the old orthodoxy. In particular, the Realists famously urged that legal rules and doctrines were sufficiently malleable and open ended to allow alternative outcomes in a wide range of cases. For example, the Realists argued that the legal rules do not provide a determinate answer to how narrowly or broadly to read individual precedents (Llewellyn 1930a, 63–66; Oliphant 1928, 72–73), or what uniting principle(s) to draw from a body of several precedents (Oliphant and Hewitt 1929, xix–xx), and that the rules do not determinately resolve which of many recognized interpretive canons, often leading to very different results, should be applied to a statute’s text (Llewellyn 1950; Radin 1930). Of course, judges still tend overwhelmingly to write their opinions as if recognized legal principles and doctrines dictate the outcome determinately (Frank 1930, 8–9; Moore and Hope 1929, 704), which might make it seem, to credulous observers, that Condition 2 substantially obtains. But in many cases, this characterization of the relationship between the rules and the decision is simply false and largely serves, consciously or not, as an ex post rationalization of a decision that really was not determinately dictated by the legal rules (Frank 1930, 100–104; Green 1928, 1021–1022; Hutcheson 1929, 279; Llewellyn 1931b, 1238–1239; Yntema 1928, 479–480).

It bears mentioning that this style of Realist critique is not a binary matter: it admits of degrees. One could, for example, take the more moderate stance that the Langdellian model would work well enough for a broad range of clear or “easy” cases, but that it is systematically inadequate for predicting outcomes in the set of cases actually litigated or (more conservatively still) those cases litigated to the appellate level (Cook 1927, 308; Llewellyn 1931b, 1239; Oliphant and Hewitt 1929, x–xi). Alternatively, one could argue that the failure is even more extensive than that (Frank 1930, 151). The Realists’ perceptions did seem to vary on this point, and they were not always perfectly clear about where they stood on it. But they were substantially united in believing that circumstances called for an alternative to a traditional legal science focused on the distillation and application of formal legal rules or doctrine.

Scientific Naturalism and the Realist Alternative

The Realists could have rested content with critiquing the traditional legal science, but in fact they did suggest an alternative – an approach to the study of law that would be more productive and methodologically sound than what came before. Far from being mere iconoclasts or skeptics, they generally shared with Langdell the desire for a genuinely scientific study of law; the crucial difference was that they parted ways with the pre-Realist orthodoxy on what a science of law would look like (Leiter 2007, 92). In order to understand the alternative Realist vision of legal science, one needs to understand something about the intellectual milieu out of which American Legal Realism emerged. As Edward Purcell has chronicled (1973, 74–94) and as Brian Leiter summarizes:

[T]he Realists…came out of the intellectual culture of the 1920s and 1930s in the United States…in which natural science was considered the paradigm of all genuine knowledge; in which science was distinguished by its methods (e.g. observation, empirical testing); and in which the social sciences aimed to emulate the methods and successes of the natural sciences. Any plausible account of Realist jurisprudence can not lose sight of this intellectual background. (2007, 18)

The relevant outlook is best captured by the label “scientific naturalism” or just “naturalism” for short. And what is naturalism? In the broadest sense, naturalism simply involves a favorable attitude toward the methods and ethos of natural science, combined with a dimmer view of other modes of inquiry or purported sources of knowledge, such as religious revelation, authority, armchair appeals to intuition or “common sense,” and so on. Sometimes naturalism also manifests in more robust substantive or ontological, as opposed to merely methodological or epistemological, views – for example, the physicalist thesis that “all things that exist in this world are bits of matter and structures aggregated out of bits of matter” (Kim 2005, 149–150). This aspect of naturalism, however, is less important for present purposes. The key point is that the Realists were naturalists in the broad, primarily methodological sense.

The most famous aspect of the affirmative Realist project, in which the Realists’ methodological naturalism is particularly evident, is their proposal for how to gain a better understanding of how courts actually decide cases: namely, by pursuing an empirical study of judicial behavior modeled on the social sciences (Leiter 2007, 40; Purcell 1973, 86–87). For the Realists, “Langdell’s ideal of the legal scholar” was to be replaced, at least in significant measure, by the “social scientist law professor,” who was “to set law in the frame of anthropology, economics, political science, psychology, and sociology” (Fisher et al. 1993, 234). This, for the Realists, would help bring the study of law, whose relatively “backward” or “medieval” state they frequently lamented (Cohen 1935, 830; Llewellyn 1931a, 89; Oliphant and Hewitt ix–x; Yntema 1928, 474), into the modern era. But how exactly would this proposal for an empirical, naturalistic science of law move from skeletal theory to concrete practice, and what would a mature Realist science of law look like? Of course, lawyers – at least good ones – already adjust their predictions about how judges will act on the basis of informal psychological and sociological posits, e.g., that a judge who is a conservative former prosecutor is more likely to decide against criminal defendants. This kind of reasoning may be broadly continuous with empirical social science and does represent a kind of inchoate Realism. But such haphazardly developed instincts lack the systematicity or rigor that would characterize a full-fledged Realist science of law (Cohen 1935, 845–846; Leiter 2007, 55–56). How much better than this could a true naturalized science of law be, and what would be the best way of developing it?

Perhaps unsurprisingly given the difficulty of answering such questions from the proverbial armchair, the Realists were not all of one mind on such matters (Leiter 2007, 25–30). At one extreme, Jerome Frank hypothesized that the decisions of courts were often affected by very particular and, indeed, idiosyncratic aspects of judges’ personalities. For example, Frank suggested that a judge might, due to specific past experiences, have strong negative or positive reactions to “blonde women, or men with beards, or Southerners…or plumbers” (1930, 106). Such idiosyncratic biases, Frank thought, could often be an outcome-determinative factor in litigation. A different view, to which a larger group of Realists adhered (Leiter 2007, 29), emphasized that there were – contra Frank’s apparent view – “significant, predictable, social determinants that govern the course of judicial decision” (Cohen 1935, 843). As Felix Cohen put it, echoing a catchphrase of Frank’s, “Judges are human”; nonetheless, “they are a peculiar breed of humans, selected to a type and held to service under a potent system of governmental controls” (1935, 843). In support of the view that judicial decisionmaking is a less chaotic phenomenon than Frank’s perspective would imply, this latter group could point out that “actual experience does reveal a significant body of predictable uniformity in the behavior of courts” (Cohen 1935, 843). For example, Herman Oliphant observed that promises not to compete seemed generally to be enforced where the seller of a business promised not to compete with the buyer, but not where an employee had promised not to compete with an employer after the employment had ended (1928, 159) – a pattern that might not be determinately dictated by legal doctrine, but which could hardly be expected to arise if judges were commonly influenced in outcome-determinative ways by factors as particular and seemingly random as affective responses to the hair color, accent, or demeanor of parties or counsel.

The stakes of this debate for the future of a naturalistic, predictive science of judicial decisionmaking were (and are) considerable. If Frank was correct that “the ultimately important influences in the decisions of any judge are the most obscure, and are the least easily discoverable – by any one but the judge himself” (1930, 114), and also that these influences vary widely from judge to judge, this would at least appear to present a serious practical difficulty for the would-be scientist of judicial decisionmaking. Perhaps the external obscurity of the relevant psychological factors could be overcome if judges willingly underwent psychoanalysis and provided “detailed autobiographies” and/or “opinions annotated…with elaborate explorations of the background factors…which swayed” their decisions (Frank 1930, 114–115). But even so, the variability of the causally important psychological factors would remain a practical impediment to a useful predictive science of judicial decisionmaking: insight into a given judge’s psychology might help with the task of predicting that specific judge’s future decisions, but it would not likely help one understand other judges, whose decisions would be the product of quite different blends of psychological biases and quirks.

On the other hand, insofar as the more optimistic Realists were correct that judges’ decisions could be predicted on the basis of externally transparent psychosocial factors common to the judiciary in general (or even to sizeable and identifiable subsets thereof), it would be possible, as Leiter puts it, to formulate “lawful (or at least lawlike) predictive generalizations about patterns of decision” (2007, 26–27). In the end, it is hard to adjudicate this disagreement between Frank and the more mainstream Realist voices; really, it may in significant part be a matter of emphasis. It is, however, important to note that even Frank looked to (then) modern psychology to understand judicial decisionmaking. He, too, was a methodological naturalist; it is just that he played up the chaotic and idiosyncratic aspects of judging, leading him to pessimistic conclusions about the feasibility of rendering adjudication manageably predictable.

Questioning the Realist Program

One of the most well-known critiques of the American Legal Realists is that, for all their insistent advocacy of an empirical, naturalistic science of law, they never truly delivered on the promise: insofar as they tried to practice what they preached, the results were generally not stellar and did not really lead anywhere (Posner 1995, 393–394). Even if that assessment is correct, however, one cannot conclude that the basic Realist proposal itself was unsound or insufficiently motivated (Leiter 2007, 54–55). Later generations have taken up social-scientific studies of adjudication very much in the spirit of American Legal Realism (Miles and Sunstein 2008), and although it is beyond the scope of the present discussion to evaluate the progress of such research, it is not out of the question that further efforts based on newer and sounder social-scientific frameworks will vindicate the Realists’ hope for successful predictive models of judicial decisionmaking based on empirical social-scientific methods.

A more distinctively philosophical worry about the Realists’ naturalization project is the following. Would an empirical, predictive inquiry into the causes (or, for that matter, the effects) of judicial or other official action actually constitute a science of law at all? It would, to be sure, illuminate important aspects of legal institutions. But generally we think of law itself as consisting, as Morris Cohen put it, of “norms which regulate, rather than…uniformities which describe, human conduct” (1931, 358). Whatever else Langdell got wrong, one might argue, he at least seemed on the right track, from a metaphysical or conceptual perspective, when he identified “principles or doctrines” – rules, in short – as the proper object of a science of law. Now it is true that the Realists were famously (or infamously) prone to embracing so-called predictive accounts of law (Cohen 1933, 11–14; Cook 1924, 465; Frank 1930, 46; Llewellyn 1930a, 3), typically variations on Oliver Wendell Holmes, Jr.’s famous dictum, “The prophecies of what the courts will do in fact, and nothing more pretentious, are what I mean by the law” (Holmes 1897, 461). Given that way of speaking, it is trivially true that an empirical predictive science of judicial decisionmaking is a science of law. The problem, however, is that this usage of “law” is jarringly out of alignment with ordinary linguistic practice, a fact made obvious if we observe that when judges ask what the law is, they are not trying to predict their own actions (Dickinson 1931, 843–844; Hart 1959, 237). Are the Realists, then, open to the accusation that they were just changing the subject – in effect urging that it would be useful for legal scholars and law students to study something other than (or in addition to) the law?

A few Realists do seem to have underappreciated how revisionary their favored way of using the word “law” was (notably Walter Wheeler Cook (1924, 475–476) and possibly, though to a lesser extent, Felix Cohen (1933, 12)), but it is important to understand that the Realists were generally not much concerned with adhering to ordinary usage. The Realist literature is, in fact, littered with caveats about the fruitlessness of debating the “correct” definitions of words like “law” (Cohen 1935, 835; Green 1928, 1014; Llewellyn 1930b, 431–433), so the accusation that the Realists used words like “law” in nonstandard ways would seem largely to miss the point. The important thing was that Langdellian legal science purported to teach lawyers-in-training what they would need to know for the practice of law, and the Realists had a proposal for an improved means to that end. In that sense, the science they proposed was a science of law, and that was what mattered to the Realists.

Even if one accepts this line of reasoning, however, there is another familiar criticism of the affirmative Realist program, one perhaps more in keeping with the Realists’ pragmatic temperament. Although it is important for practicing lawyers to be able to predict official action, is it not also important to know how judges should decide cases? Law schools, after all, train future judges, too. Under the (nonstandard) predictive definition of “law,” of course, the latter problem would concern what the law should be, not what it is; but whatever one wants to call it, the fact remains that any full replacement for Langdellian legal science would have to address this normative problem as well (Kronman 1993, 199). For although Langdellian legal science purported to be a study of what the law (considered as a body of prescriptive rules) is, not what it should be, it could still claim to provide judges with a reliable normative guide to decisionmaking. After all, supposing the legal rules dictate a determinate outcome in substantially all cases, and given the hardly unusual supposition that judges should follow the legal rules, all that remains to answer the question of how judges should decide cases is to learn what the legal rules are and how to apply them – precisely the knowledge Langdell’s legal science purported to provide.

Now, if we accept the Realist thesis of substantial indeterminacy in legal doctrine, this simply will not do. So long as “ought” implies “can,” then insofar as the legal rules are indeterminate, it cannot be that judges ought to (if only because they are unable to) “just follow the rules” when deciding cases. Thus, the Realists’ indeterminacy thesis would undermine the utility of doctrinal knowledge not only for purposes of predicting judicial action but also for purposes of providing a normative guide for judges themselves. But while an empirical science of judicial decisionmaking can plausibly replace Langdellian legal science qua guide to external prediction of judicial action, it is hard to see how it could fully replace Langdellian legal science qua normative guide to adjudication. Certainly, a judge does not figure out how to decide by trying to predict his or her own future actions. And although one could credibly argue that trial court judges and intermediate appellate judges ought always to decide in the way most likely to survive further review – in effect trying to predict their superiors’ reactions – it is more difficult to extend this logic to judges who sit at the peak of a given jurisdiction’s appellate hierarchy. (Any effort to do so would, at minimum, put one well outside the mainstream of American legal thought.)

Can the Realists fairly be accused of sidelining normative problems in their zeal to replace “the traditional approach to law in terms of…official rules” with “a new approach to law as a social science” (Llewellyn 1931a, 119)? Llewellyn, having detected this accusation in Roscoe Pound’s famous 1931 critique of Realism, replied that the Realists advocated only a “temporary divorce of Is and Ought” (1931b,1236). One’s views on what courts ought to do must not be allowed to distort one’s inquiry into what courts will in fact do (Llewellyn 1931a, 98–101, b, 1236–1237), but that does not mean one must not ultimately pass judgment on value questions: “observation material is not enough to live by, nor yet to do by” (Llewellyn 1931a, 100). And it is plainly true that the Realists themselves, far from being simply disinterested observers of the legal processes they studied, had plenty of normative views regarding the law (Ursin 2012). The Realists were generally political progressives, more or less consequentialist in outlook, and their advocacy of more social science in legal scholarship and education was not unrelated to their hopes for reform (Purcell 1973, 93): a common Realist theme was that in an increasingly complex society, judges and other lawmakers needed more social-scientific training in order to set good policy (Cook 1931, 109n31; Llewellyn 1931b, 1248–1250; Oliphant 1928, 159–161).

Still, there was a detectable discomfort on the part of many Realists toward robust, systematic normative theorizing, at least some of which surely derived from characteristically naturalist concerns about the subjectivity of ethics – as Llewellyn put it, that when “we move into…value judgments we desert entirely the solid sphere of objective observation, of possible agreement among all normal trained observers, and enter the airy sphere of individual ideals and subjectivity” (1931a, 100). Be that as it may, it is important not to overindulge the stereotype of the American Legal Realists as obsessed with value-neutral empirical study at the expense of systematic inquiry into questions of value (Fried 1998, 14). Felix Cohen provides a useful counterexample: far from being averse to normative theorizing, he lamented what he saw as a widespread reluctance among legal scholars to consciously engage in systematic ethical inquiry (Cohen 1934, 35) and wrote extensively about both ethical and metaethical theory himself (Cohen 1933).


The American Realist perspective on matters of legal science involved both a critical agenda and a constructive one, both of which have had substantial effects on American legal thought. Most legal thinkers in the United States today accept the Realist critique of the old orthodoxy at least in a moderate-to-mild form. The constructive Realist proposal for bringing social-scientific methods to bear on the questions that matter most to the legal profession has also had considerable influence, even if the hope for a systematic predictive science of judicial decisionmaking remains largely unfulfilled. But while most contemporary legal thinkers accept some Realist ideas, lawyers and legal scholars continue to differ on many of the subjects addressed above, including the extent of (in)determinacy in legal doctrine, the methods to be used in legal scholarship, and the kind of knowledge that is most useful for legal practice. The American Realists’ ideas about the law and its study accordingly remain subjects not only of historical interest but also contemporary relevance and controversy.



  1. Beale JH (1916) A treatise on the conflict of laws, vol 1, pt 1. Harvard University Press, CambridgeCrossRefGoogle Scholar
  2. Cohen MR (1931) Justice Holmes and the nature of law. Colum L Rev 31:352–367CrossRefGoogle Scholar
  3. Cohen FS (1933) Ethical systems and legal ideals. Falcon, New YorkGoogle Scholar
  4. Cohen FS (1934) Modern ethics and the law. Brook L Rev 4:33–50Google Scholar
  5. Cohen FS (1935) Transcendental nonsense and the functional approach. Colum L Rev 35:809–849CrossRefGoogle Scholar
  6. Cook WW (1924) The logical and legal bases of the conflict of laws. Yale L J 33:457–488CrossRefGoogle Scholar
  7. Cook WW (1927) Scientific method and the law. A.B.A. J 13:303–309Google Scholar
  8. Cook WW (1931) Law and the modern mind: a symposium (part III, legal logic). Colum L Rev 31:108–115Google Scholar
  9. Dickinson J (1931) Legal rules: their function in the process of decision. U Pa L Rev 79:833–868Google Scholar
  10. Fisher WW, Horwitz MJ, Reed TA (eds) (1993) American legal realism. Oxford University Press, New YorkGoogle Scholar
  11. Frank J (1930) Law and the modern mind. Brentano’s, New YorkGoogle Scholar
  12. Frank J (1931) Are judges human?, pt 1. U Pa L Rev 80:17–53Google Scholar
  13. Frank J (1932) What courts do in fact, pt 1. Ill L Rev 26:645–666Google Scholar
  14. Fried BH (1998) The progressive assault on laissez faire. Harvard University Press, CambridgeGoogle Scholar
  15. Green L (1928) The duty problem in negligence cases. Colum L Rev 28:1014–1045CrossRefGoogle Scholar
  16. Grey TC (1983) Langdell’s orthodoxy. U Pitt L Rev 45:1–53Google Scholar
  17. Grey TC (1996) Modern American legal thought. Review of Duxbury N (1995) patterns of American jurisprudence. Yale L J 106:493–517CrossRefGoogle Scholar
  18. Hart HLA (1959) Scandinavian realism. Camb L J 1959:233–240CrossRefGoogle Scholar
  19. Holmes OW (1897) The path of the law. Harv L Rev 10:457–478CrossRefGoogle Scholar
  20. Hutcheson JC (1929) The judgment intuitive: the function of the “hunch” in judicial decision. Cornell L Q 14:274–288Google Scholar
  21. Kim J (2005) Physicalism, or something near enough. Princeton University Press, PrincetonGoogle Scholar
  22. Kimball BA (2004) The Langdell problem: historicizing the century of historiography, 1906–2000s. Law Hist Rev 22:277–337CrossRefGoogle Scholar
  23. Kimball BA (2009) The inception of modern professional education: C. C. Langdell, 1826–1906. University of North Carolina Press, Chapel HillCrossRefGoogle Scholar
  24. Kronman AT (1993) The lost lawyer. Harvard University Press, CambridgeGoogle Scholar
  25. Langdell CC (1879) A selection of cases on the law of contracts, 2nd edn. Little, Brown, BostonGoogle Scholar
  26. Langdell CC (1880) A summary of the law of contracts, 2nd edn. Little, Brown, BostonGoogle Scholar
  27. Langdell CC (1887) Harvard celebration speeches: Professor Langdell. L Q Rev 3:123–125Google Scholar
  28. Leiter B (2007) Naturalizing jurisprudence. Oxford University Press, OxfordCrossRefGoogle Scholar
  29. Llewellyn KN (1930a) The bramble bush. Columbia University School of Law, New YorkGoogle Scholar
  30. Llewellyn KN (1930b) A realistic jurisprudence – the next step. Colum L Rev 30:431–465CrossRefGoogle Scholar
  31. Llewellyn KN (1931a) Legal tradition and social science method – a realist’s critique. In: Essays on research in the social sciences. Brookings Institution, Washington, DC, pp 89–120Google Scholar
  32. Llewellyn KN (1931b) Some realism about realism – responding to dean pound. Harv L Rev 44:1222–1264CrossRefGoogle Scholar
  33. Llewellyn KN (1950) Remarks on the theory of appellate decision and the rules or canons about how statutes are to be construed. Vand L Rev 3:395–406Google Scholar
  34. Miles TJ, Sunstein CR (2008) The new legal realism. U Chi L Rev 75:831–851Google Scholar
  35. Moore U, Hope TS (1929) An institutional approach to the law of commercial banking. Yale L J 38:703–719CrossRefGoogle Scholar
  36. Oliphant H (1928) A return to stare decisis. A.B.A. J 14:71–76, 107, 159–162Google Scholar
  37. Oliphant H, Hewitt A (1929) Introduction. In: Rueff J, From the physical to the social sciences (trans: Green H). Johns Hopkins Press, Baltimore, pp ix–xxxiiGoogle Scholar
  38. Posner RA (1995) Overcoming law. Harvard University Press, CambridgeGoogle Scholar
  39. Purcell EA (1973) The crisis of democratic theory. University Press of Kentucky, LexingtonGoogle Scholar
  40. Radin M (1925) The theory of judicial decision: or how judges think. A.B.A. J 11:357–362Google Scholar
  41. Radin M (1930) Statutory interpretation. Harv L Rev 43:863–885CrossRefGoogle Scholar
  42. Schauer F (2013) Legal realism untamed. Tex L Rev 91:749–780Google Scholar
  43. Sebok AJ (1995) Misunderstanding positivism. Mich L Rev 93:2054–2132CrossRefGoogle Scholar
  44. Ursin E (2012) The missing normative dimension in Brian Leiter’s “reconstructed” legal realism. San Diego L Rev 49:1–35Google Scholar
  45. Wambaugh E (1906) Professor Langdell – a view of his career. Harv L Rev 20:1–4Google Scholar
  46. Yntema HE (1928) The hornbook method and the conflict of laws. Yale L J 37:468–483CrossRefGoogle Scholar

Copyright information

© Springer Nature B.V. 2018

Authors and Affiliations

  1. 1.University of Chicago Law SchoolChicagoUSA

Section editors and affiliations

  • Pierluigi Chiassoni
    • 1
  1. 1.Università degli Studi di GenovaGenovaItaly