Subtitling and translation play such vital roles in the global film industry, it's something of an enigma that those responsible behind the scenes go so unsung. Most people recognize that bad subtitles can ruin a film - one only has to look at the Hong Kong DVD releases of Ozu's films about five years ago as an example of how much can go lost in translation. At the same time, how many of us come out of the latest arthouse smash saying "Wow, what a great film! The subtitles really captured the emotion and nuance of the dialogue, and so succinctly and eloquently too...? We only really notice subtitles when they let us down.

Bad subtitling can add an unintended comic dimension to an otherwise serious work. Most of us have our own favourite examples of bad subbing. Mine is that wonderful scene from the original Hong Kong VCD of Herman Yau's delirious The Ebola Syndrome (1996), in which one of the characters returns home to find his wife being brutally raped, and screams at the assailant "You bully my wife and I'll bully you!" It's the ludicrous schism between the violence portrayed on screen and the rather timorous language here that makes us laugh, but at least the message comes across.

More sinister is when the subtitles convey something completely different from what the characters are expressing, when subtle nuances are either exaggerated or lost altogether, or in short, when bad translation not only hinders meaning, but severely betrays intent. It happens with rather more regularity than one might think, and not only in cinema. I vividly recall an experience in France back in 1992 while staying with the family of a former girlfriend and watching a news report featuring Prince Charles giving a heartfelt plea against the overregulation of local food markets by European Union bureaucrats. Using the examples of cheese, he stated plainly that with too much centralised control one would never maintain the variety in local food production that made, for example, a trip to French supermarkets for British people such a blissful experience. Ironically, while he waxed poetically over "the heady aroma of a ripe Camembert and the sturdy Roquefort", I could see from my hosts' faces that the subtitles were telling a different story. "What an idiot!" they screamed, "He's insulting our cheese!" - and as every European knows, the worst thing you can ever do to a French person is mock their cheese. Back in those sensitive early years of the EU, it's amazing to think that such simple newsroom negligence had the potential to kick off the next cross-Channel skirmish ("You are English. You don't understand. You have only your Cheddar," I remember was the general direction in which the conversation subsequently headed). I learned a valuable lesson during this seminal spell abroad, which is how much power these unseen mediators of the media have in damaging relations between different cultures.

Following hot on the heels of Forest of Pressure, Nornes's last book about the documentary-making collective Ogawa Productions, Cinema Babel comes chock-full of such examples, to provide an expansive, insightful, and thoroughly entertaining picture of how translation practices over the past hundred years have helped and hindered the global flow of moving images. Nornes actually prefers the term "trafficking" over superficially similar words such as distribution, movement or circulation, because it draws attention to the powerful role the translation process has in regulating what gets shown where and how it is perceived by viewers not familiar with the source language or culture.

From its inception, cinema was the original borderless medium. A century before the internet provided instant access to a seemingly infinite pool of text and images, it presented its spectators with vivid visual evidence of what was happening on the other side of the world. One of the first things the Lumière Brothers did after showcasing their Cinématographe in Paris was dispatch a number of delegates to the far flung reaches of the globe to capture scenes of daily life from these distant lands to be shown to spectators back at home. This is how François-Constant Girel wound up in Tokyo in 1897, directly instigating the birth of Japan's own film industry. It is also how various countries began to present their own public face to other nations. In 1907, for example, only a third of the films distributed in America were actually produced there.

Early films were silent of course, but in cinema's formative years, they were often presented with an off-screen interpreter to explain to viewers exactly what it was they were seeing. In Japan, this led to the institution of the benshi film narrator, a phenomenon treated in considerable detail within these pages. The benshi in Japan were often considered a greater draw for audiences than the film itself, but their dominant presence during screenings, during which they peppered their descriptions of what was happening on screen with their own interpretations and ideas, simultaneously tamed and contaminated the original texts by bringing them under the control of a single voice, divesting "the power of interpretation from the audience, in effect becoming the exemplary spectator that individuals were asked to follow."

The increased adoption of text intertitles saw the practice of an accompanying narration dying out in the West at a fairly early stage, aside from in a few diasporic communities. For a film to be shown in territories speaking a different language than the one in which the film was produced, all that was necessary now was to snip out and replace the onscreen text with the target language equivalent - although as Nornes points out, this wasn't quite the straightforward process one might imagine. Intertitles however shifted the onus of interpretation from the exhibitor to the distributor, granting a greater degree of centralised control over the meaning of the films. Moreover, the style and purpose of the written intertitle itself underwent considerable evolution over the years, becoming an evermore integral part of the filmic text. One of Alfred Hitchcock's first jobs in the film industry was preparing intertitles for silent films. As Nornes informs us, "aside from the possibilities offered by tone and color, Hitchcock wrote at the time that illustrated intertitles do far more than vivify the action they are embedded in... Hitchcock is cognizant of cinema's temporal aspect, and how intertitles interact with live action in time."

Indeed, these pages on early film trafficking present a far more complicated side to the nascent years of this global medium than one might expect. It is often casually assumed that it was the arrival of sound that brought about Hollywood's dominance of the global movie market, but in reality the developments brought about by the groundbreaking popularity of The Jazz Singer initially proved a major bugbear for California's movie moguls: in the years immediately following 1928, Hollywood saw a 50% drop in foreign receipts. The shock of the new provided by the arrival of the talkies was considerable, with Nornes's citation of one Japanese pundit, Masanobu Takeyama, that "Thanks to the talkie, the internationality of cinema has vanished", echoing similar sentiments felt by those in the industries of many a country.

Film producers worldwide reacted in various ways. For a while London's Elstree studios made Multiple Language Versions (MLVs) of a number of titles, effectively re-filming scenes in French, German, and English to ensure the widest possible distribution, although this practice added considerably to production costs. The ensuing decades continued to see a long process of debate regarding the virtues of subbing and dubbing. "Certainly, too, I prefer to see Miss Bardot's bosom without typographic obscuration," wrote the New Statesman critic Stanley Kauffman in 1960, "But the subtitle gives you the gist of the dialogue and allows you to enjoy the actor's whole performance."

Subtitling and dubbing are not the only subjects that come under scrutiny within this discussion of the cross-cultural trafficking of the moving image. The role of the onstage interpreter at international film festivals, the influence of (mis-) translated books on film theory on a national cinematic culture, and the invisible behind-the-scenes wranglings of international co-productions also come up for discussion. It is no surprise, given Nornes's pedigree, that many of the anecdotes and examples in the book come from the world of Japanese cinema, including an extended look at the battles which raged during the production of Tora! Tora! Tora! (1970). This co-production represents something of a historical market point, in that not only did it symbolise a new period of accordance between America and Japan 25 years after the end of the Pacific War, it "was notable as one of Hollywood's first efforts to let foreign characters speak (subtitled) foreign languages for scenes on the other side of the front lines." Alas, thanks to the efforts of "the translator at the heart of all the production's troubles, Aoyagi Tetsuro," Kurosawa was almost driven insane by trying to assert his creative will over the film's American producers, and was subsequently replaced by Toshio Masuda and Kinji Fukasaku. Reading Nornes's account, it's a wonder the film ever made it to the screen at all, but it certainly wasn't quite the conciliatory monument originally intended.

There are other equally entertaining examples: the faithful translation of the colourful language used throughout Stanley Kubrick's Full Metal Jacket (1986) by Masato Harada, one of the few truly bilingual directors in the Japanese film industry, and a figure who also worked extensively on Japanese-language versions of Hollywood productions, proved scandalous for Japanese audiences, with its "unflinching translation of the fearful obscenity pouring from the drill instructor's mouth". Then there's the account of how the botched job of a certain "celebrity" translator forced the Japanese distributors of Lord of the Rings to pull the film from circulation after a couple of days, following a less than positive response from local viewers unfamiliar with Tolkien's stories; and of how Takeshi Kitano's rise to international prominence might have been nipped in the bud had it not been for a vigilant Italian translator who sensibly decided to leave certain parts of the director's acceptance speech for the Golden Lion awarded to Hana-bi at Venice in 1997 untranslated.

Nornes also draws attention to the broader ramifications of translation outside of the Japanese market. The international success of Crouching Tiger, Hidden Dragon (2000) points to the fact that American audiences may not be as resistant to subtitled films as has been historically assumed. (Jonathan Rosenbaum, in his superb book Movie Wars: How Hollywood and the Media Limit What Movies We Can See published in 2000, makes a similar point about how few people were put off by the subtitled sequences in Schindler's List and Dances with Wolves.) Perhaps the issue lies with the quality of translation? As he points out, few Hollywood directors bother to oversee the translation process of their work, seemingly unconcerned with how it is received in countries outside of America, and "Hollywood filmmakers spend more on catering than translation." And then there's the "neo-imperial translation project" of the multi-language Genesis file so beloved of DVD producers, where English forms the centralized source language, and translations from one non-English language to another face considerable risk of becoming a series of Chinese whispers of (mis-) translations of (mis-) translations.

The main question viewers always ask themselves when watching a foreign film, is how easy is it to translate jokes, puns, sarcasm, obscenities, and double entendres into a different language in a medium "where border crossings involve altercation and contamination, prophylaxis and destruction". Nornes puts forward his case for an "abusive translation", which sticks to the spirit of the original text rather than the letter, citing Neil Gaiman's script for Ghibli's Princess Mononoke as one example, and makes a passionate and surprisingly convincing defence of dubbing, by way of an in-depth look at the Czech voice-acting industry.

Like Nornes's previous books, Cinema Babel does have its more densely-worded and theoretical passages, which may trouble non-academics, but all in all it's a thoroughly entertaining and enlightening read. For those working in film translation, filmmakers cognizant that there is a far wider market for their work than a merely domestic one, and film viewers curious about the nuts and bolts of the process of subtitling and dubbing, this makes for essential reading.