The Language of Thieves is a historical inquiry, family memoir, and meditation on the idiosyncrasies of language. If the parts cohered, the book would be smashing. They don’t, which is frustrating, but Martin Puchner’s book has its merits.

The origins of The Language of Thieves are situated in Nuremberg, where Puchner grew up. As a very young boy in the 1970s he watched his mother charitably feed “strange figures” who spoke a “strange dialect.” When he was older his father told him that the perplexing language was called “Rotwelsch” and the strangers were “travelers, . . . people of the road, escaping to nowhere.” Martin’s Uncle Günter, his father’s brother, explained why the travelers knew that they could obtain food at Martin’s house. In the house’s foundation stone he pointed out a circle with a cross inside, a symbol understood by travelers to mean that “there was bread to be had here.” As Martin eventually discovered, there were “dozens of such signs,” peculiar images — zinken they were called in Rotwelsch — that complemented the peculiar spoken language. The signs, he says, “turned . . . my world into a labyrinth of mysterious symbols. . . . Finding them became an obsession.” Puchner became hooked not just on zinken but, in due course, on Rotwelsch as a whole.

As it turned out, another member of the Puchner family had also been drawn to Rotwelsch — Martin’s contemptible paternal grandfather, Karl. When Martin was in his twenties, he was in Harvard’s Widener Library and had a sudden urge to see if any writing by Karl, who had been a librarian, historian, and archivist, was stored there. What he found shocked him: a 1934 German scholarly journal that contained an article by Karl that flagrantly pandered to the new Nazi government. It concentrated on the need for the state to prevent Jews from using Aryan names, but it also expressed hatred for Rotwelsch: “For my grandfather, there was one language mixture that was even more disgusting than Yiddish: Rotwelsch.” Rotwelsch contained a number of Hebrew and Yiddish words, and since Karl believed that Rotwelsch speakers were criminals (only some were), all Jews had to be criminals too (if the reasoning escapes you, it escapes Martin Puchner too — and me). Karl was also convinced that Rotwelsch had insinuated itself into German, perverting it. Martin, who had always found his late grandfather amiable, eventually learned that Karl wasn’t merely an opportunistic Nazi but a dedicated one. “My grandfather,” Martin sadly concludes, “made me see that, in a strange, twisted way, Rotwelsch was perfect for anti-Semites. It seemed to confirm their most irrational ideas by connecting language, race, and criminality. . . . Jews were polluters of language and language thieves.” As a result of Karl’s vileness, however, Martin found a calling: “For me, reckoning with the Holo­caust became the main reason for studying history.”

Günter, the man who had deciphered zinken for his young nephew, Martin, was also captivated by Rotwelsch and zinken. Perhaps “captivated” isn’t strong enough. After Günter’s death, his widow showed Martin boxes of material that his uncle had amassed on Rotwelsch and zinken: books, manuscripts, translations, police files, Romani and Yiddish dictionaries, pamphlets, field notes, etc. — a veritable treasure trove of information. Martin pored over his uncle’s lucubrations, and the cache sparked his own further investigations. And so he learned (as will his readers) a great deal about a most obscure, but provocative, curiosity.

Rotwelsch seemingly was mostly confined to Central Europe and France. (Similar “underground” languages existed — and exist — throughout the world.) It incorporated, to a certain extent, the vocabulary of regions where it took root (and gleefully changed meanings; some of the spinoffs from Yiddish are rather droll), but its grammatical scaffolding was derived from “various German dialects.” Notwithstanding its Yiddish and Hebrew elements (contributed, mostly, by roving Jewish tradesmen and, yes, criminals), “there was nothing specifically Jewish about Rotwelsch speakers,” writes Puchner. “The confusion of Rotwelsch and Yiddish was primarily the work of anti-Semites for whom speaking a Jewish-inflected German and speaking a thieves’ language was pretty much the same thing.” Who spoke Rotwelsch? The speakers included “escaped convicts, runaway apprentices, deserters, itinerant peddlers, tramps, professional thieves, beggars, hoboes, journeymen, knife grinders, tinkers, migrants, and anyone at odds with the authorities and without a fixed address.” No incubators of persiflage there, I imagine (nor of bombast, which is a good thing).

Rotwelsch, maintains Puchner, “was purely a spoken language (except for a few dozen zinken),” and it goes way back. “The earliest sighting of the term dated to ce 1250” (the author provides no further details); it meant, at the time, “deceptions.” It was probably Martin Luther, in the early 16th century, who popularized the word. But not favorably. Luther believed that tramps, beggars, and the like were a danger to decent people, and therefore he also execrated their distinct language: As Puchner puts it, it was, for Luther, “the incomprehensible cant of beggars.” Moreover, by the time Luther turned his attention to these matters, he disdained Jews too, and since he had decided that (here Puchner quotes Luther directly) “the Rotwelsch language comes from the Jews,” he felt, Puchner says, that Rotwelsch was “hostile to Christendom.” Over the centuries, not only were the speakers of Rotwelsch persecuted, but the language was stigmatized. Therefore, unsurprisingly, the police, through the centuries, became the main — the only — keepers of written documentation on this perceived disreputable language: “Producing a record of this language . . . was precisely the way in which [the authorities] wanted to eliminate it. Writing things down, fixing words and names, registering vagrants; all these were connected.” Rotwelsch had its proponents: Kafka was one. The language intrigued him, and although he didn’t use it in his own work, “he recognized Rotwelsch as a cultural phenomenon to be cherished, and as a resource for theater and literature.”

The Nazis almost destroyed Rotwelsch: It “barely survived the systematic incarceration and elimination of vagrants by the Nazi regime.” (Hitler had been a vagrant in Vienna as a young man. One presumes there was a connection between this and the Nazis’ war against vagrants.) But Puchner discovered that Rotwelsch is (and zinken are) still a marginal, sub rosa means of communication in Switzerland. Also, zinken appeared in an episode of Mad Men. It doesn’t get more illustrious than that. (I jest. Incidentally, I’m surprised that Puchner fails to mention that Thomas Pynchon’s 1965 novel, The Crying of Lot 49, dabbles in zinkenlike cryptic glyptics.)

Martin Puchner is a professor of English and comparative literature at Harvard and the general editor of The Norton Anthology of World Literature. His scholarly credentials are impeccable, and he comes across as a fair and decent man. The most moving part of this book is his decision not to tell his aunt — Karl’s daughter — about her father’s villainy. After making the decision, he felt “ashamed,” but he needn’t have. He was right not to reveal something that would only torment an elderly lady.

The discussions of Rotwelsch and zinken are the best parts of the book. But Puchner’s goal was to weave the Rotwelsch material into research about, and reminiscences of, his family, and unfortunately the attempt, for the most part, doesn’t come off. Yes, the nexus between his grandfather, uncle, and Rotwelsch is relevant and thought-provoking. But there is too much extraneous family history, particularly about his mother’s antecedents.

At the end of The Language of Thieves, Puchner offers a poignant plea to the world to leave Rotwelsch alone. Part of his appeal goes, “If I have learned anything from studying Rotwelsch, it is that we, the settlers, the people with passports and states, the ones who own territories, should not seek to control this language; perhaps we shouldn’t even record it, since the history of studying Rotwelsch is so closely associated with attempts to suppress it.” At the risk of sounding churlish (or worse), I don’t think that the disappearance of Rotwelsch would be a great calamity for humanity. I advert to Puchner’s remark that Rotwelsch is an oral language that has produced no written body of work. Forgive me if I sound like a snooty mandarin, but I’m of the opinion that it is primarily in a language’s literature (broadly defined) that its mystique can become apparent — a numinous quality, if you will, that Anthony Burgess called “a region of word-magic and word-superstition.”

Considering Puchner’s intellectual bona fides, I’m surprised that he didn’t discuss the issues raised by Rotwelsch’s lack of a literary component. I, for one, wonder whether a language used mostly for communicating mundane, unadorned, diurnal information can even be considered a language. Puchner does briefly ponder the Third Reich’s exploitation of the German language — a subject that might involve venturing into Burgess’s theater of necromantic antics. Based on his research, Puchner concludes that “the Nazis didn’t construct an entirely new language; instead they twisted existing terms.” Puchner’s guide in this was the academic Victor Klemperer, who experienced Nazi misdeeds at first hand. Drawing from Klemperer, Puchner cites a number of examples, including führer (this is Puchner): “Traditionally, it had been used in all kinds of benign contexts, . . . but now Führer meant only one thing: Adolf Hitler.”

Compare this to a passage from a 1959 essay (“The Hollow Miracle”) by the late man of letters George Steiner: “The German language was not innocent of the horrors of Nazism. It is not merely that a Hitler, a Goebbels, and a Himmler happened to speak German. Nazism found in the language precisely what it needed to give voice to its savagery. Hitler heard inside his native tongue the latent hysteria, the confusion, the quality of hypnotic trance. He plunged unerringly into the undergrowth of language, into those zones of darkness and outcry which are the infancy of articulate speech, and which come before words have grown mellow and provisional to the touch of the mind.” Steiner might have been wrong (Burgess thought he was), but he didn’t hesitate to delve into what he perceived was German’s chthonic nightmare. Puchner shunned that incursion.

The Language of Thieves is worth reading; you will be in capable hands, the hands of a good writer and sensitive scholar. The book raises tantalizing, compelling issues. But unavoidably, I suspect, it suffers somewhat because it reflects the fraught nature of its torturous concerns.