The Lost Files of Sherlock Holmes

In 1989, Trip Hawkins reluctantly decided to shift Electronic Arts’s strategic focus from home computers to videogame consoles, thereby to “reach millions of customers.” That decision was reaching fruition by 1992. For the first time that year, EA’s console games outsold those they published for personal computers. The whole image of the company was changing, leaving behind the last vestiges of the high-toned “software artists” era of old in favor of something less intellectual and more visceral — something aimed at the mass market rather than a quirky elite.

Still, corporate cultures don’t change overnight, and the EA of 1992 continued to release some computer games which were more in keeping with their image of the 1980s than that of this new decade. One of the most interesting and rewarding of these aberrations — call them the product of corporate inertia — was a game called The Lost Files of Sherlock Holmes, whose origin story doesn’t exactly lead one to expect a work of brilliance but which is in fact one of the finest, most faithful interpretations of the legendary detective in the deerstalker cap ever to make its way onto a monitor screen.

The initial impetus for Lost Files was provided by an EA producer named Christopher Erhardt. After studying film and psychology at university, Erhardt joined the games industry in 1987, when he came to Infocom to become the in-house producer for their latter-day lineup of graphical games from outside developers, such as Quarterstaff, BattleTech: The Crescent Hawk’s Inception, and Arthur: The Quest for Excalibur. When Infocom was shuttered in 1989, he moved on to EA in the same role, helming a number of the early Sega Genesis games that did so much to establish the company’s new identity. His success on that front gave him a fair amount of pull, and so he pitched a pet idea of his: for a sort of computerized board game that would star Sherlock Holmes along with a rotating cast of suspects, crimes, and motives, similar to the old 221B Baker Street board game as well as a classic computer game from Accolade called Killed Until Dead. It turned out that EA’s management weren’t yet totally closed to the idea of computer games that were, as Erhardt would later put it, “unusual and not aimed at the mass market” — as long, that is, as they could be done fairly inexpensively.

In order to meet the latter condition, Erhardt enlisted a tiny Tempe, Arizona, company known as Mythos Software — not to be confused with the contemporaneous British strategy-games developer Mythos Games. This Mythos was being run by one James Ferguson, its fresh-out-of-university founder, from the basement of his parents’ house. He was trying to break into the wider world of software development that lay outside the bounds of the strictly local contracts he had fulfilled so far; his inexperience and eagerness ensured that Mythos would work cheap. And in addition to cut-rate pricing, Ferguson had another secret weapon to deploy: an artist named Scott Mavor who had a very special way with pixel graphics, a technique that EA’s in-house employees would later come to refer to as “the Mavor glow.” The highly motivated Mythos, working to Erhardt’s specifications, created a demo in less than two weeks that was impressive enough to win the project a tentative green light.

Another EA employee, a technical writer named Eric Lindstrom, saw the demo and suggested turning what had been planned as a computerized board game into a more narratively ambitious point-and-click adventure game. When Erhardt proved receptive to the suggestion, Lindstrom put together the outline of a story, “The Mystery of the Serrated Scalpel.” He told Erhardt that he knew the perfect person to bring the story to life: one of his colleagues among EA’s manual writers, a passionate Sherlock Holmes aficionado — he claimed to have read Arthur Conan Doyle’s complete canon of Holmes stories “two or three times” — named R.J. Berg.

The project’s footing inside EA was constantly uncertain. Christopher Erhardt says he “felt like I was playing the Princess Bride, and the dread pirate Roberts was coming.  It was always, ‘Yep – we may cancel it.'” But in the end the team was allowed to complete their point-and-click mystery, despite it being so utterly out of step with EA’s current strategic focus, and it was quietly released in the fall of 1992.

I find the critical dialog that followed, both in the immediate wake of Lost Files‘s release and many years later in Internet circles, to be unusually interesting. In particular, I’d like to quote at some length from Computer Gaming World‘s original review, which was written by Charles Ardai, one of the boldest and most thoughtful — and most entertaining — game reviewers of the time; this I say even as I find myself disagreeing with his conclusions far more often than not. His review opens thus:

If there is any character who has appeared in more computer games than Nintendo’s plump little goldmine, Mario, it has to be Sherlock Holmes. There have been almost a dozen Holmes-inspired games over the years, one of the best being Sherlock Holmes Consulting Detective, which is currently available in two different CD-ROM editions from ICOM. Other valiant attempts have included Imagic’s Sherlock Holmes in Another Bow, in which Holmes took a sea voyage with Gertrude Stein, Picasso, Thomas Edison, and Houdini, among others; and Infocom’s deadly serious Sherlock: Riddle of the Crown Jewels.

The difference between Holmes and Mario games, however, is that new Mario games are always coming out because the old ones sold like gangbusters, while new Sherlock Holmes games come out in spite of the fact that their predecessors sold like space heaters in the Sahara. It is noteworthy that, until ICOM, no company had ever released more than one Sherlock Holmes game, while all the Mario games come from the same source. It is also worth noting that the Holmes curse is not limited to games: the last few Holmes movies, such as Without a Clue and Young Sherlock Holmes, were not exactly box-office blockbusters.

The paradox of Sherlock Holmes can be stated so: while not that many people actually like the original Sherlock Holmes stories, everyone seems to think that everyone else adores them. Like Tarzan and Hawkeye, Holmes is a literary icon, universally known and much-beloved as a character in the abstract — not, however, as part of any single work. Finding someone who has actually read and enjoyed the writing of Edgar Rice Burroughs, James Fenimore Cooper, or Arthur Conan Doyle requires the patience of Diogenes. Most people know the character from television and the movies, at best; at worst, from reviews of television shows and movies they never bothered to see.

So, why do new Holmes adaptations surface with such regularity? Because the character is already famous and the material is in the public domain (thereby mitigating the requisite licensing fees associated with famous characters of more recent vintage. Batman or Indiana Jones, for instance.) Another answer is that Sherlock Holmes is seen as bridging the gap between entertainment and literature. Game companies presumably hope to cash in on the recognition factor and have some of the character’s ponderous respectability rub off on their product. They also figure that they can’t go wrong basing their games on a body of work that has endured for almost a century.

Unfortunately for them, they are wrong. There are only so many copies of a game that one can sell to members of the Baker Street Irregulars (the world’s largest and best-known Sherlock Holmes fan club), and a vogue for Victoriana has never really caught on among the rest of the game-buying population. The result is that, while Holmes games have been good, bad, and indifferent, their success has been uniformly mediocre.

This delightfully cynical opening gambit is so elegantly put together that one almost hesitates to puncture its cogency with facts. Sadly, though, puncture we must. While there were certainly Sherlock Holmes games released prior to Lost Files that flopped, there’s no evidence to suggest that this was the fault of the famous detective with his name on the box, and plenty of evidence to the contrary: that his name could, under the right circumstances, deliver at least a modest sales boost. In addition to the Sherlock Holmes Consulting Detective CD-ROM productions, a counter-example to Ardai’s thesis that’s so huge even he has to acknowledge it — the first volume of that series sold over 1 million units — there’s also the Melbourne House text adventure Sherlock; that game, the hotly anticipated followup to the bestselling-text-adventure-of-all-time The Hobbit, likely sold well over 100,000 units in its own right in the much smaller market of the Europe of 1984. Even Infocom’s Riddle of the Crown Jewels, while by no means a smash hit, sold significantly better than usual for an Infocom game in the sunset of the company’s text-only era. (Nor would I describe that game as “deadly serious” — I could go with “respectful” at most — but that’s perhaps picking nits.)

Still, setting aside those inconvenient details, it’s worth considering this broader question of just why there have been so many Sherlock Holmes games over the years. Certainly the character doesn’t have the same immediate appeal with the traditional gaming demographic as heavyweight properties like Star Wars and Star Trek, Frodo Baggins and Indiana Jones — or, for that matter, the born-in-a-videogame Super Mario. The reason for Sherlock’s ubiquity in the face of his more limited appeal is, of course, crystal clear, as Ardai recognizes: he’s in the public domain, meaning anyone who wishes to can make a Sherlock Holmes game at any time without paying anyone. ^[1]There have been occasional questions about the extent to which Sherlock Holmes and his supporting cast truly are outside all bounds of copyright, usually predicated on the fact that the final dozen stories were published in the 1920s, the beginning of the modern copyright era, and thus remain protected. R.J. Berg remembers giving “two copies of the game and a really trivial amount of money” to Arthur Conan Doyle’s aged granddaughter, just to head off any trouble on that front. When a sequel to Lost Files of Sherlock Holmes was published in 1996, no permission whatsoever was sought or demanded.

As such, Holmes occupies a nearly unique position in our culture. He’s one of the last great fictional icons, historically speaking, who’s so blessedly free of intellectual-property restrictions. Absolutely everyone, whether they’ve ever read a story or seen a film featuring him or not, knows him. The only characters with a remotely similar degree of recognizability who postdate him are Dracula, the Wizard of Oz, and Peter Pan — and neither of the latter two at least presents writers with quite the same temptation to tell new story after story after story.

As is noted in Lost Files‘s manual, Sherlock Holmes has become such an indelible part of our cultural memory that when we see him we experience a sort of vicarious nostalgia for a London none of us ever knew: “Gas lamps, the sound of horses’ hooves, steam locomotives, and romantic street cries. And then there is the atmosphere of that cozy room in Baker Street: Holmes in his armchair before a roaring coal fire, legs stretched out before him, listening with Dr. Watson to yet another bizarre story.” One might say that Sherlock Holmes gets the chronological balance just right, managing to feel both comfortably, nostalgically traditional and yet also relevant and relatable. In contrast to the Victorian scenery around him, his point of view as a character feels essentially modern, applicable to modern modes of storytelling. I’m not sure that any other fictional character combines this quality to quite the same extent with a freedom from copyright lawyers. These factors have fostered an entire creative subculture of Sherlockia which spans the landscape of modern media, dwarfing Arthur Conan Doyle’s canonical four novels and 56 short stories by multiple orders of magnitude.

The relative modernity of Sherlock Holmes is especially important in the context of interactive adaptations. The player of any narrative-driven game needs a frame of reference — needs to understand what’s expected of her in the role she’s expected to play. Thankfully, the divide between Sherlock Holmes and the likes of C.S.I. is a matter of technology rather than philosophy; Sherlock too solves crimes through rationality, combining physical evidence, eyewitness and suspect interviews, and logical deduction to reach a conclusion. Other legendary characters don’t share our modern mindset; it’s much more difficult for the player to step into the role of an ancient Greek hero who solves problems by sacrificing to the gods or an Arthurian knight who views every event as a crucible of personal honor. (Anyone doubtful of Sherlock Holmes’s efficacy in commercial computer games should have a look at the dire commercial history of Arthurian games.)

With so much material to make sense of, post-Doyle adventures of Sherlock Holmes get sorted on the basis of various criteria. One of these is revisionism versus faithfulness. While some adaptations go so far as to transport Sherlock and his cronies hook, line, and sinker into our own times, others make a virtue out of hewing steadfastly to the character and setting described by Arthur Conan Doyle. This spirit of Sherlockian fundamentalism, if you will, is just one more facet of our long cultural dialog around the detective, usually manifesting as a reactionary return to the roots when other recent interpretations are judged to have wandered too far afield.

No matter how much the Sherlockian fundamentalists kick and scream, however, the fact remains that the Sherlock Holmes of the popular imagination has long since become a pastiche of interpretations reflecting changing social mores and cultural priorities. That’s fair enough in itself — it’s much of the reason why Doyle’s timeless sleuth remains so timeless — but it does make it all too easy to lose sight of Holmes and Watson as originally conceived in the stories. Just to cite the most obvious example: Holmes’s famous deerstalker cap is never mentioned in the text of the tales, and only appeared on a few occasions in the illustrations that originally accompanied them. The deerstalker became such an iconic part of the character only after it was sported by the actor Basil Rathbone as an item of daily wear — an odd choice for the urban Holmes, given that it was, as the name would imply, a piece of hunting apparel normally worn by sporting gentlemen in the countryside — in a long series of films, beginning with The Hound of the Baskervilles in 1939.

Although Lost Files doesn’t go so far as to forgo the deerstalker — there are, after all, limits to these things — it does generally try to take its cue from the original stories rather than the patchwork of interpretations that followed them. Berg:

I definitely aimed for Holmesian authenticity. I’d like to think that, if he were alive, Doyle would like the game. After all, the characters of Holmes and Watson have been manipulated quite a bit by the various media they’ve appeared in, especially the films. For example, the Watson of Lost Files is definitely Doyle’s Watson, competent and intelligent, rather than the bumbling character portrayed in many of the movies. I also wanted to retain Holmes’s peculiar personality. He’s really not that likable a character; he’s arrogant, a misogynist, and extremely smug.

This spirit of authenticity extends to the game’s portrayal of Victorian London. There are, I’ve always thought, two tiers when it comes to realistic portrayals of real places in fiction. Authors on the second tier have done a whole lot of earnest research into their subject, and they’re very eager to show it all to you, filling your head with explicit descriptions of things which a person who actually lived in that setting would never think twice about, so ingrained are they in daily existence. Authors on the top tier, by contrast, have seemingly absorbed the setting through their pores, and write stories that effortlessly evoke it without beating you over the head with all the book research they did to reach this point of imaginative mastery.

Lost Files largely meets the latter bar as it sends you around to the jewelers and tobacconists, theaters and pubs, opulent mansions and squalid tenements of fin-de-siècle London. The details are there for when you need them or decide to go looking for them; just try mousing around the interior of 221B Baker Street. (“A typical sitting-room chair. The sheen of its wine-red velveteen covering shows that it is well-used. A dark purple silk dressing gown with a rolled collar is carelessly crumpled on the seat and the antimacassar requires changing.”) More impressive, though, is the way that the game just breathes its setting in that subtle way that can only be achieved by a writer with both a lighter touch and countless hours of immersion in the period at his command. For example Berg spent time reading Charles Dickens as well as Arthur Conan Doyle in order to capture the subtle rhythms of Victorian English in his conversations. This version of Holmes’s London isn’t the frozen-in-amber museum exhibit it sometimes comes off as in other works of Sherlockia. “We wanted a dirty game,” says Eric Lindstrom. “We wanted people to feel that people were burning coal, that they could see who was walking in the streets. Just as it was in London at the time.”

There is, however, one important exception to the game’s rule of faithfulness to the original stories: Lost Files presents a mystery that the reader can actually solve. In light of the place Holmes holds in our cultural memory as the ultimate detective, one of the great ironies of Doyle’s stories is that they really aren’t very good mysteries at all by the standard of later mystery fiction — a standard which holds a good mystery to be an implicit contest between writer and reader, in which the reader is presented with all the clues and challenged to solve the case before the writer’s detective does so. Doyle’s stories cheat egregiously by this standard, hiding vital evidence from the reader, and often positing a case’s solution on a chain of conjecture that’s nowhere near as ironclad as the great detective presents it to be. Eric Lindstrom:

The [original] stories do not work the way we are used to today. They are not whodunnits; whodunnits only became popular later. Readers have virtually no way of finding out who the culprit is. Sometimes the offender does not even appear in the plot. These are adventure stories narrated from the perspective of Dr. Watson.

For obvious reasons, Lost Files can’t get away with being faithful to this aspect of the Sherlock Holmes tradition. And so the mystery it presents is straight out of Arthur Conan Doyle — except that it plays fair. Notably, you play as Holmes himself, not, as in the original stories, as Watson. Thus you know what Holmes knows, and the game can’t pull the dirty trick on you, even if it wanted to, of hiding information until the big reveal at the end. Many other works of Sherlockia — even the otherwise traditionalist ones — adapt the same approach, responding to our post-nineteenth-century perception of what a good mystery story should be.

And make no mistake: “The Case of the Serrated Scalpel” is a very good mystery indeed. I hesitate to spoil your pleasure in it by saying too much, and so will only state that what begins as the apparently random murder of an actress in an alley behind the Regency Theatre — perhaps by Jack the Ripper, leaving Whitechapel and trying his hand in the posher environs of Mayfair? — keeps expanding in scope, encompassing more deaths and more and more Powerful People with Secrets to Keep. As I played, I was excited every time I made a breakthrough. Even better, I felt like a detective, to perhaps a greater extent than in any computer game I’ve ever played. Among games in general, I can only compare the feeling of solving this mystery to that of tackling some of the more satisfying cases in the Sherlock Holmes Consulting Detective tabletop game.

Part of the reason the mystery comes together so well is just down to good adventure-game design principles, of the sort which eluded so many other contemporary practitioners of the genre. Berg:

The idea was to produce a game that was different from existing adventures, which I frankly felt were often tedious. We wanted to eliminate the elements that tend to detract from the reality of the experience — things like having to die in order to learn some crucial information, constantly having to re-cover the same territory, and the tendency to simply pick up and use every object you encounter. We wanted to give players a deeper experience.

So, there are none of the dreaded adventure-game dead ends in Lost Files. More interestingly, the design does, as Berg alludes above, mostly eschew the typical use-unlikely-object-in-unlikely-place model of gameplay. Tellingly, the few places where it fails to do so are the weakest parts of the game.

As I’ve noted before, the classic approach to the adventure game, as a series of physical puzzles to solve, can be hugely entertaining, but it almost inevitably pushes a game toward comedy, often in spite of its designers’ best intentions. Most of us have played alleged interactive mysteries that leave you forever messing about with slider puzzles and trivial practical problems of the sort that any real detective would solve in five minutes, just by calling for backup. In Infocom’s Sherlock: Riddle of the Crown Jewels, for example, you learn that a stolen ruby is hidden in the eye of the statue of Lord Nelson on top of Nelson’s Column, and then get to spend the next little while trying to get a pigeon to fetch it for you instead of, you know, just telling Inspector Lestrade to send out a work crew. Lost Files does its level best to resist the siren call of the trivial puzzle, and, with only occasional exceptions, it succeeds. Thereby is the game freed to become one of the best interactive invocations of a classic mystery story ever. You spend your time collecting and examining physical evidence, interviewing suspects, and piecing together the crime’s logic, not solving arbitrary road-block puzzles. Lost Files is one of the few ostensibly serious adventure games of its era which manages to maintain the appropriate gravitas throughout, without any jarring breaks in tone.

This isn’t to say that it’s po-faced or entirely without humorous notes; the writing is a consistent delight, filled with incisive descriptions and flashes of dry wit, subtle in all the ways most computer-game writing is not. Consider, for example, this description of a fussy jeweler: “The proprietor is a stern-looking woman, cordial more through effort than personality. She frequently stares at the cleaning girl who tidies the floor, to make sure she is still hard at work.” Yes, this character is a type more than a personality — but how deftly is that type conveyed! In two sentences, we come to know this woman. I’d go so far as to call R.J. Berg’s prose on the whole better than that of the rather stolid Arthur Conan Doyle, who tended to bloviate on a bit too much in that all too typical Victorian style.

The fine writing lends the game a rare quality that seems doubly incongruous when one considers the time in which it was created, when multimedia was all the rage and everyone was rushing to embrace voice acting and “interactive movies.” Ditto the company which published it, who were pushing aggressively toward the forefront of the new mass-media-oriented approach to games. In spite of all that, darned if Lost Files doesn’t feel downright literary — thoughtful, measured, intelligent, a game to take in slowly over a cup of tea. Further enhancing the effect is its most unique single technical feature: everything you do in the game is meticulously recorded in an in-game journal kept by the indefatigable Dr. Watson. The journal will run into the hundreds of onscreen “pages” by the time you’re all done. It reads surprisingly well too; it’s not inconceivable to imagine printing it out — the handy option to print it or save it to a file is provided — and giving it to someone else to read with pleasure. That’s a high standard indeed, one which vanishingly few games could meet. But I think that The Lost Files of Sherlock Holmes just about manages it.

Having given so much praise to Lindstrom and Berg’s design and writing, I have to give due credit as well to Mythos Software’s efforts to bring it all to life. The interface of Lost Files is thoroughly refined and pleasant to work with, a remarkable achievement considering that this was the first point-and-click graphic adventure to be made by everyone involved. An optional but extremely handy hotspot finder minimizes the burden of pixel hunting, and the interface is full of other thoughtful touches, like a default action that is attached to each object; this saves you more often than not from having to make two clicks to carry out an action.

Finally, praise must be given to Scott Mavor’s “Mavor glow” graphics as well. To minimize the jagged edges typical of pictures drawn in the fairly low resolution of 256-color VGA graphics, Mavor avoided sharp shifts in color from pixel to pixel. Instead he blended his edges together gradually, creating a lovely, painterly effect that does indeed almost seem to glow. Scott’s mother Elinor Mavor, who worked with him to finish up the art in the latter stages of the project: ^[2]Scott Mavor died of cancer in 2008

Working with just 256 colors, Scott showed me how he created graduating palettes of each one, which allowed him to do what he called “getting rid of the dots” in each scene. To further mute the pixels, he kept the colors on the darker side, which also enhanced the Victorian mood.

Weaving the illusion of continuous-tone artwork with all those little “dots” made us buggy-eyed after a long day’s work. One night, I woke up, went into the bathroom, turned on the light, and the world just pixelated in front of me. Scary imprints on my retinas had followed me away from the computer monitor, rendering my vision as a pointillistic painting à la George Seurat.

While the graphics of its contemporaries pop out at you with bright, bold colors, the palette of Lost Files of Sherlock Holmes smacks more of the “brown sauce” of the old masters — murky, mysterious, not initially jaw-dropping but totally in keeping with the mood of the script. As you, playing the diligent detective, begin to scan them carefully, the pictures reveal more and more details of the sort that are all too easy to overlook at a quick glance. It makes for an unusually mature aesthetic statement, and a look that can be mistaken for that of no other game.

Given all its strengths, I find it surprising that Lost Files has gotten more than its share of critical flak over the years. I have a theory as to why that should be, but before I get to that I’ll let one of the naysayers share his point of view. Even after admitting that the story is “a ripping yarn,” the graphics atmospheric, the period details correct, and the writing very good, Charles Ardai concludes his review thusly:

Don’t get me wrong: the dialogue is well-written, the choices are entertaining, and in most cases the actions the game requires the player to perform are very interesting. The story is good and the game is a pleasure to watch. Yet, that is what one does — watch.

This game wants, more than anything in the world, to be a Sherlock Holmes movie. Though it would be a very good one if it were, it is not. Therefore, it is deeply and resoundingly unsatisfying. The plot unfolds quite well, with plenty of twists, but the player has no more control over it than he would if he were reading a novel. The player is, at best, like an actor in a play. Unfortunately, said actor has not been given a copy of the script. He has to hit his marks and say his lines by figuring out the cues given by the other characters and reading his lines off the computer equivalent of cue cards.

If this is what one wants — a fine Sherlock Holmes pastiche played out on the computer screen, with the player nominally putting the lead character through his paces — fine. “The Case of the Serrated Scalpel” delivers all that one could hope for in that vein. If one wants a game — an interactive experience in which one’s decisions have an effect on what happens — this piece of software is likely to disappoint.

The excellent German podcast Stay Forever criticized the game along similar — albeit milder — lines in 2012. And in his mostly glowing 2018 review of the game for The Adventure Gamer joint-blogging project, Joe Pranevich as well noted a certain distancing effect, which he said made him feel not so much like he was playing Sherlock Holmes and solving a mystery as watching Sherlock do the solving. The mystery, he further notes — correctly — can for the most part be solved by brute force by the patient but obtuse player, simply by picking every single conversation option when talking to every single character and showing each of them every single object you’ve collected.

At the extreme, criticisms like these would seem to encroach on the territory staked out by the noted adventure-game-hater Chris Crawford, who insists that the entire genre is a lie because it cannot offer the player the ability to do anything she wants whenever she wants. I generally find such complaints to be a colossal bore, premised on a misunderstanding of what people who enjoy adventure games find most enjoyable about them in the first place. But I do find it intriguing that these sorts of complaints keep turning up so often in the case of this specific game, and that they’re sometimes voiced even by critics generally friendly to the genre. My theory is that the mystery of Lost Files may be just a little bit too good: it’s just enticing enough, and just satisfying enough to slowly uncover, that it falls into an uncanny valley between playing along as Sherlock Holmes and actually being Sherlock Holmes.

But of course, playing any form of interactive fiction must be an imaginative act on the part of the player, who must be willing to embrace the story being offered and look past the jagged edges of interactivity. Certainly Lost Files is no less interactive than most adventure games, and it offers rich rewards that few can match if you’re willing to not brute-force your way through it, to think about and really engage with its mystery. It truly is a game to luxuriate in and savor like a good novel. In that spirit, I have one final theory to offer you: I think this particular graphic adventure may be especially appealing to fans of textual interactive fiction. Given its care for the written word and the slow-build craftsmanship of its plotting, it reminds me more of a classic Infocom game than most of the other, flashier graphic adventures that jostled with it for space on store shelves in the early 1990s.

Which brings me in my usual roundabout fashion to the final surprising twist in this very surprising game’s history. After its release by a highly skeptical EA, its sales were underwhelming, just as everyone had been telling Christopher Erhardt they would be all along. But then, over a period of months and years, the game just kept on selling at the same slow but steady clip. It seemed that computer-owning Sherlock Holmes aficionados weren’t the types to rush out and buy games when they were hot. Yet said aficionados apparently did exist, and they apparently found the notion of a Sherlock Holmes adventure game intriguing when they finally got around to it. (Somehow this scenario fits in with every stereotype I carry around in my head about the typical Sherlock Holmes fan.) Lost Files‘s sales eventually topped the magical 100,000-unit mark that separated a hit from an also-ran in the computer-games industry of the early- and mid-1990s.

Lost Files of Sherlock Holmes may not have challenged the likes of John Madden Football in the sales sweepstakes, but it did make EA money, and some inside the company did notice. In 1994, they released a version for the 3DO multimedia console. For the sake of trendiness, this version added voice acting and inserted filmed footage of actors into the conversation scenes, replacing the lovely hand-drawn portraits in the original game and doing it no new aesthetic favors in the process. In 1996, with the original still selling tolerably well, most of the old team got back together for a belated sequel — The Lost Files of Sherlock Holmes: Case of the Rose Tattoo — that no one would ever have dreamed they would be making a couple of years before.

But then, almost everything about the story of Lost Files is unlikely, from EA of all companies deciding to make it — or, perhaps better said, deciding to allow it to be made — to a bunch of first-time adventure developers managing to put everything together so much better than many established adventure-game specialists were doing at the time. And how incredibly lucky for everyone involved that such a Sherlock Holmes devotee as R.J. Berg should have been kicking around writing manuals for EA, just waiting for an opportunity like this one to show his chops. I’ve written about four Sherlock Holmes games now in the course of this long-running history of computer gaming — yet another measure of the character’s cultural ubiquity! — and this one nudges out Riddle of the Crown Jewels to become the best one yet. It just goes to show that, no matter how much one attempts to systematize the process, much of the art and craft of making games comes down to happy accidents.

(Sources: Compute! of April 1993 and June 1993; Computer Gaming World of February 1993; Questbusters of September 1988 and December 1992; Electronic Games of February 1993. Online sources include Elinor Mavor’s remembrances of the making Lost Files of Sherlock Holmes, the comprehensive Game Nostalgia page on the game, the Stay Forever podcast episode devoted to the game, Joe Pranevich’s playthrough for The Adventure Gamer, the archived version of the old Mythos Software homepage, and Jason Scott’s “Infocom Cabinet” of vintage documents.

Feel free to download Lost Files of Sherlock Holmes from right here, in a format designed to be as easy as possible to get running under your platform’s version of DOSBox or using ScummVM.)