There’s an old joke about scientific progress: that science advances one funeral at a time. The same might be said about judicial philosophies. Some judges might be persuaded to change their views over time, but it is just as likely, if not more, that change occurs through a changing of the guard. So it was with the Second Circuit in the 1960s. The Second Circuit had had a remarkably stable bench during the 1940s, when Learned Hand was the chief judge. Four of them, Hand, Hand’s cousin Augustus, Harrie Chase, and Thomas Swan served together in active or senior status for twenty-five years, from 1929 to 1954. The remaining two, Charles Clark and Jerome Frank, were with them from 1940 on.
But within six years in the 1950s, the active bench of the Second Circuit experienced nearly a complete turnover, except for Clark. The new judges, who included Henry Friendly, J. Edward Lumbard, Irving Kaufman, and Thurgood Marshall, obviously had different educational and professional experiences from those of the judges they replaced. What truly distinguished the new group, however, is that they had a very different approach to judicial decisionmaking, and in particular the proper role of discretion. No longer were the Second Circuit judges comfortable with leaving important substantive decisions on the merits of a claim to case-by-case equitable balancing. In the 1960s, the Second Circuit began crafting multi-part tests to replace the vague standards that had come before, to force lower courts and later panels to elaborate the reasons for their decisions. Whether they consciously subscribed to it or not, the new judges were heavily influenced by Legal Process ideology.
The judges of Learned Hand’s generation, as well as those who came before, were perfectly comfortable rendering decisions that explained the basis for the outcome in a sentence or two. Opinions were short, typically only a few pages or less. For example, in Dymow v. Bolton, the court declared, after briefly reviewing the similarities between works, “This incomplete skeleton the two plays have in common, but it is with real difficulty that the flesh and blood, the incidental, yet essential, adornment and trimming, of the plays can be cut away to show similarity between a few bones. This difficulty is fatal to plaintiff’s case ….” In Fleischer Studios, Inc. v. Ralph A. Freundlich, Inc., the result went the other way: “What the appellant constructed is recognizable by an ordinary observer as having been taken from the copyrighted source. Such is an infringement. The essential characteristics of appellees’ copyrighted character are reproduced.”
Nor was such terseness limited to substantial similarity. A few years ago I tried to track down the origin of the rule that games are not copyrightable. The early cases were inscrutable, saying such things as, without authority, “In the conventional laws or rules of a game, … there can be no literary property susceptible of copyright.” The early cases on indirect liability were similarly devoid of explication. Take Justice Holmes’s explanation in Kalem Co. v. Harper Brothers, for why the defendant was indirectly liable for infringement: “If the defendant did not contribute to the infringement, it is impossible to do so except by taking part in the final act. It is liable on principles recognized in every part of the law.” The Second Circuit in Gross v. Van Dyk Gravure Co.: “Why all who unite in an infringement are not, under the statute, liable for the damages sustained by plaintiff, we are unable to see.” The Seventh Circuit in Dreamland Ball Room, one of the so-called “dance hall” cases: “The authorities are, we believe, unanimous in holding that the owner of a dance hall at whose place copyrighted musical compositions are played in violation of the rights of the copyright holder is liable, if the playing be for the profit of the proprietor of the dance hall.” (A Harvard Law Review Note the following year described this as “unanimous, though scanty, opinion.”)
The seven judges who served on the Second Circuit for more than a year in the 1930s graduated law school between between 1894 and 1913. Most of them came to the bench straight from private practice; two were law professors. Only Chase had worked, briefly, as a prosecutor, and only Frank had any significant regulatory experience. It may not have been a coincidence that Frank wrote Arnstein, which went further than any opinion before in delineating the bifurcated infringement test.
The new cadre of Second Circuit judges that took their seats in the 1950s and 1960s had vastly different experiences that those of the Hand era. For the most part, they graduated law school in the 1920s and early 1930s, the heyday of legal realism in the academy, although none were realists. They then worked in private practice at a time when the administrative state was coalescing. A quarter had extensive experience with regulatory agencies. Nearly half had served as prosecutors or defense attorneys. Two had been legislators.
When this new group of judges took the bench starting in the mid-50s, they were surprised by the state of copyright law. Theirs was not a typical revolution. Most intellectual revolts follow a pattern: the ideological framework of a previous generation, itself revolutionary in its time, slowly reveals its gaps and flaws. As circumstances change, those flaws become less and less tolerable, until a new generation overthrows the old framework completely and replaces it with their own.
At least with respect to copyright law, however, the reaction of the new judges was not frustration or anger, but perplexity. As I’ve noted before, in three significant opinions between 1957 and 1963, three different newly appointed judges expressed amazement that some issue relating to liability for infringement was unresolved by both the courts and the statute, fifty years after its adoption. Judge Irving Kaufman almost sounded weary by the third opinion in the sequence, Shapiro, Bernstein & Co. v. H. L. Green Co.:
This action for copyright infringement presents us with a picture all too familiar in copyright litigation: a legal problem vexing in its difficulty, a dearth of squarely applicable precedents, a business setting so common that the dearth of precedents seems inexplicable, and an almost complete absence of guidance from the terms of the Copyright Act.
What these judges appear to have been expecting was a guide for their decisions, such as a multi-factor test or elements of a claim, of the kind that the Restatements had adopted twenty or thirty years before – just as they were graduating law school. In particular, they appear to have consciously or unconsciously to have adopted the precepts of the Legal Process school.
The Legal Process school originated at Harvard in the 1950s as a reaction to legal realism, which after World War II was increasingly viewed as problematic. The legal process theorists, along with others, were disturbed by legal realists’ support for judicial discretion over not just the outcome of cases, but the content of the law. Over the course of a decade or so in the 1950s and 1960s, legal process scholars, in articles and in course materials for an eponymous class at Harvard, identified an alternative approach to judicial decisionmaking. Two components are key for our purposes: institutional settlement and reasoned elaboration.
“Institutional settlement” refers to the notion that a decision by a state actor – whether a judge, regulator, legislator, or executive – gains legitimacy as law and precedent if it is decided according to the procedures applicable to the decisionmaker’s institution. In other words, each institutional actor has an obligation, if they want their decisions to be respected, to stay within the limits of their institutional role. Legal process ideology thus took a sharp turn away from recognizing judges as policymakers, even in common law subjects; that was a job for legislators and regulators. While it didn’t invent the concept, the Legal Process School marked a forceful return of the idea of judges as strictly apolitical resolvers of conflicts between rules created by others.
The second principle was “reasoned elaboration,” the idea that valid and authoritative decisions by courts could not simply be delivered by fiat, but must be based on an explanation of how the decision is consistent with the body of existing law and “best serves the principles and policies it expresses.”
The Legal Process School is not particularly famous for its influence on lower courts, or on private law subjects like copyright. Most often it is associated with the course taught by Henry Hart and Albert Sacks, and with its strong criticism of the Warren Court in a series of articles and lectures at Harvard. Both of those efforts famously fizzled out; the course materials were never published, and legal process scholars were never able to identify how, as a procedural matter, a court was to decide between competing substantive starting points for the analysis. The epitome of the legal process critique of the Warren Court came in Henry Hart’s 1963 Holmes Lectures, in which he was to lay out his solution to the problem of deciding between conflicting constitutional principles, when midway through his second lecture he declared that the approach he had planned would not work, and sat down.
But while the Legal Process School did not have a direct impact on Supreme Court decisionmaking, its core ideas seem to have had an enormous impact on lower courts, including the Second Circuit. The exact causal mechanism is a bit unclear. Judge Henry Friendly read the Legal Process course materials prior to taking the bench, and pronounced himself extremely impressed with it. But other Second Circuit judges may not have even heard of the Legal Process. Regardless, they seem as a group to have internalized its core concepts of institutional settlement and reasoned elaboration, which may have been circulating in the legal culture for both practicing lawyers and law professors at Harvard and Yale to pick up.
The difference in approach can be seen in the debate between Learned Hand, writing his last copyright opinion, and Henry Friendly, writing his first, in Peter Pan Fabrics. Friendly dissented, but not from Hand’s description of the standard for improper appropriation as depending on the aesthetic appeal to the ordinary observer, but rather from Hand’s treatment of the statutory requirement of notice. Section 10 of the 1909 Act required a copyright notice to be “affixed to each copy … offered for sale.” While the bolts of cloth Peter Pan sold to clothing manufacturers had notices along the edge, the fabric as incorporated into garments and sold to consumers typically did not. Applying “relentless literalism,” Hand wrote, it would appear that the sale of the dresses was “publication” without notice, and therefore brought the fabric into the public domain.
But Hand was not willing to be so relentless. “[I]t is a commonplace that a literal interpretation of the words of a statute is not always a safe guide to its meaning.” Instead, the court put the onus on the defendant to show that the notice could have been incorporated directly into the design “without impairing its market value,” even though such an affirmative defense is nowhere provided in the statute.
Judge Friendly dissented. “I could reconcile the majority’s result with the language of § 10 if, but only if, there were clear evidence that the dominant intention of Congress was to afford the widest possible copyright protection whereas the notice requirement was deemed formal or at least secondary. I find nothing to support such a stratified reading of § 10.” That was true even though Friendly recognized that “it may not be practicable to affix the notice to an inside seam on every repetition of the design” and there was a good argument that such notice was “a useless formality” for dress manufacturers. Friendly felt his hands were tied by the institutional settlement that had left such decisions to Congress:
[A]s was held by the Supreme Court in Dejonge, it is not for the courts to say that something less than the statutory requirement will serve…. Perhaps my brothers are right in thinking that Congress wished literal compliance with § 10 to be excused under such circumstances as here; but the voice so audible to them is silent to me.
A reluctance to fill copyright’s gaps was only one of the effects of legal process ideology. A more significant impact resulted from judges’ attempts to spell out guidelines for the doctrines that copyright had previously left to judicial discretion: indirect liability, fair use, and, most relevant for this series, substantial similarity. We’ll see what happened when the Legal Process School met “aesthetic appeal” in the next post.