Wednesday, October 6, 2010

Retrospective: Murder in Harmony

While it's long been my contention that most old school adventure modules were what I'd call "location-based" scenarios, I recently realized that I neglected to include another possibility: "problem-based" scenarios. Scenarios of this sort don't necessarily focus on a single location (though they may), but on a single problem that the adventurers must resolve. The problem can range from the simple -- put an end to the orc raids on civilized lands by slaying their leader -- to the complex -- investigate the disappearance of a human colony on the edge of space -- but the problem plays a similar role to the location in location-based scenarios, namely providing a backdrop against which the characters perform their actions.

A good example of a problem-based scenario is a murder mystery. In murder mystery modules, the characters are faced with uncovering who killed someone and why through the uncovering of clues that allow them to piece together the truth. Adventures of this type are particularly difficult to pull off well, both in terms of presentation and in terms of execution, which is why modules like 1982's Murder in Harmony stand out even after nearly three decades. Written by Mark Acres for Gangbusters, Murder in Harmony concerns the murder of Arthur Overton, a wealthy musician and president of the Amalgamated Musicians' Union. Overton was shot dead during a brief blackout at a cocktail party at Harmony Manor, Overton's stately home and it's up to the PCs to solve this mystery.

I remember Murder in Harmony for several reasons, chief among them being that it was the first truly successful murder mystery module I ever ran. I'd run murder mystery adventures before, but they were always of my own devising, so I knew their intricacies backward and forward. One of the dangers of problem-based adventures, as I stated above, is presentation. How do you present the referee with all the information he needs in order to run the module effectively, especially if there are a large number of details, many of which are red herrings or at least ancillary to the central problem of the module? Murder in Harmony has no simple solution to this, instead opting for a series of grouped sections -- chronology of events, keyed encounters, testimony, physical evidence, financial records, wiretaps, etc. -- all of which the referee is expected to know well, or at least well enough that he can easily find the information he needs when the characters start poking around.

The other thing I love about Murder in Harmony isn't really about the module itself so much as about the type of old school play it (and Gangbusters itself) exemplifies. All of the clues that help the characters resolve the central problem of the module can only be found by looking in the right places or asking the right questions. There are no mentions of "a DC 15 Gather Information check reveals ..." or "Succeed at an Observation roll to notice ..." in this module. Instead, the text assumes that the players, through their characters, will try and think things through on their own, collecting information by visiting the crime scene, interviewing witnesses, and generally employing basic investigative techniques to amass enough clues to point them to other clues that might enable them to resolve the central problem of the module. Murder in Harmony is thus very much about challenging the player, not the character, which I think is an important feature of old school gaming.

Make no mistake: Murder in Harmony is a challenging module, both to run and to participate in as a player. The referee needs to know a lot of information, as the text is jam-packed with details, a great many of which, as I noted, are irrelevant to the mystery of Overton's murder. Players must rely on their own wits and intuitions in order to know where to go, what to look for, and whom to interview if they have any hope of solving this mystery. Now, truth be told, the mystery isn't all that complex or difficult to unravel; in some respects, it's fairly conventional, especially when looked at within the context of crime stories and films from the era. Nevertheless, there are no freebies in this module. The players are largely thrown on their own resources to find the clues that can establish the truth beyond a reasonable doubt and bring the perpetrator to justice.

Running Murder in Harmony when I was a teenager was an important part of my gaming education, which probably explains why I retain such a fondness for it even now. I'm sure that it could probably be better presented than it is. Much of its text consists of large "info-dumps" of names, places, times, and other details that the referee must become comfortable with and that means lengthy preparation beforehand. I doubt I could easily run a module of this sort nowadays, not without a more streamlined and user friendly presentation. But, back when I was a kid, I had all the time in the world and I had no difficulties in spending hours getting everything just right so that I could have fun with my friends as they tried to solve the mystery of Arthur Overton's murder.

My lasting impression of Murder in Harmony, though, is that it presents a style of play I still find compelling, one where the player is every bit as important as his character, probably moreso. I remember people who insisted that this module's mystery was "unsolvable." That's far from literally true, but it does speak, I think, to the fact that, if the players don't make the right choices, they won't be able to unravel the mystery this module presents. Personally, I think that's how it should be. After all, what's the point of a mystery if it doesn't require that -- and I mean you, not your character -- try to puzzle it out for yourself?


  1. I played D&D back in the day, but I really got into playing RPGs through my local comic shop. We played the hell out of DC Heroes. From the moment it came out until about the early 90's we played just about every module that they released. Most of them were just what you described. You had to totally "crack a case" as I would sum it up. I loved the aspect of the game that was all about solving the mystery. It is my favorite RPG of all time.

  2. I think you're right that there is great potential in innovating presentation of information in something like this. Perhaps a visual timeline might convey more detail than text entries by time, etc (don't know how this module handles that).

    Also intrigued by the possibility of problem based adventures besides murder-mysteries. Thanks for sharing.

  3. I think to properly run a mystery adventure the DM has to be willing to let the PC fail. The problem i've found both playing and running mystery adventures is that is rarely satisfing to either party. As a player I just get frustrated that I am not smart enough and as a DM I get frustrated that the players can't grasp the what I see as obvious clues I have laid out for them especialy when they start looking at it like its my fault.

    If I turn back from the dungeon because I don't think my P.C. is tough enough to take on a monster I'm fine, but I have played endless hours of mystery adventures going nowhere to try to prove I am smart.

    I love mystery adventures when they work. but they can lead to hours of sitting around a table talking in circles.

  4. Also intrigued by the possibility of problem based adventures besides murder-mysteries.

    They're much harder to do, because, unlike murder mystery investigation, the skills needed to succeed are not (generally) those possessed by most players. I mean, finding a cure to an alien plague is something that I think could be done as a problem in an adventure that doesn't rely on either a simple series of dice rolls to achieve is possible, but it'd take care to present well without assuming all the players have knowledge of epidemiology.

  5. I think to properly run a mystery adventure the DM has to be willing to let the PC fail.

    This is the crux of the matter.

  6. Presentation of clues is always difficult with some players.
    I GMed many "misteries", but never until I knew how well would react players to failure. Many players hate it, and cannot cope with even a small failure (that could open up new adventure possibilities), much less if "it's their fault" not having seen the obvious. And that is more common with the younger generations, at least around here.

    As for preparation, since I tend to be a very "sandboxing" GM (I don't like that term, and it's the way I always have GMed), the only thing I always prepare is a mostly ironclad timetable of events.
    Mostly, because things can change depending on players actions, of course.

    Maybe that's why I never liked dungeoncrawls :P

  7. An interesting modern twist on this type of game is Gumshoe. The rules are pretty far from anything that people would hail as "old school," but for a mystery scenario (or Cthulhu investigation, as in the Gumshoe-derived Trail of Cthulhu), they're excellent. The fundamental idea of Gumshoe is that investigators are good at investigating; if they look for clues where there's a clue to be found, they find it. The game will never derail because players can't find clues (unless they're really thick players). The challenge and enjoyment comes from separating the rule clues from the red herrings and then figuring out what the clues mean in order to solve the mystery. All of that is done by the players, not their characters. It's quite a lot of fun. As was GangBusters, of course. It's one of my favorites.


  8. I highly recommend GURPS Mysteries for anyone interested in this sort of game. It covers pretty much every aspect of running any sort of mystery scenario (although murders get the lion's share of the focus, in keeping with the mystery genre as a whole).

    Amazingly, that it took until 2005 for such an accessory to be released!

  9. I second Will's recommendation of GURPS Mysteries. Definitely one of the best RPG supplements ever. GURPS Mysteries contains tons of useful advice. For example, it points out that a mystery that looks simple on paper (in a written adventure or novel) will often be surprisingly challenging for players in actual gameplay. An adventure that seems far too simple to you, the GM, is likely to feel just about right to your players. So the GM should keep it simple, give LOTS of relevant clues (multiple paths to the solution) and not throw in too many random and unimportant details or try to be too clever. One plot twist or brilliant deduction per adventure is usually enough.

    Unfortunately, the default scenario implied throughout GURPS Mysteries tends to be "New School" -- structured around a series of "scenes" with a "path of least resistance" built in, as well as a large section devoted to skills checks. I think that the mystery RPG genre is one area that is begging for some old school, location-based innovations.

    the DM has to be willing to let the PC fail.

    Failure should be an option, but this is a tricky issue. Even if players have a great time exploring the world, if they don't actually solve the murder, there will be a tendency to feel that the whole adventure was a "failure." Just imagine if you borrowed an Agatha Christie novel from the library and, as you were just about to finish it, you discovered that the last few pages were missing. No matter how well-written the book was, no matter how many colorful characters or entertaining twists you had encountered along the way, most mystery readers would not be satisfied with the whole experience without the solution.

    GURPS Mysteries also makes a useful distinction between two basic types of mystery adventure designs: the Jigsaw Puzzle and the Ball of Twine. With the Ball of Twine, the detectives start with a single fact or thing to investigate. That fact leads them to the next fact, which leads to the next fact, and so on. The twine unravels as a series of "scenes" which leads to the uncovering and confronting of the adversary. The Jigsaw Puzzle is more of a location-based sandbox: a large puzzle is scattered on the table. The pieces are the clues which the detectives can easily acquire by exploring the crime scene, questioning witnesses, etc. A potential problem here is that once the pieces are on the table, if the investigators can put the pieces together, the mystery is solved. If they don’t, it isn’t. A hybrid of the Jigsaw Puzzle and the Ball of Twine often works well.

    I'd love to see an Old School supplement a la GURPS Mysteries, but with the emphasis on sandbox play rather than a trail of "scenes" and skill rolls.

  10. An interesting modern twist on this type of game is Gumshoe.

    I know lots of people think well of it, but I'm not keen on GUMSHOE. There's something about its approach that just rubs me the wrong way. I'm probably not giving it a fair shake, I know, but there it is.

  11. I've only encountered GUMSHOE through Trail of Cthulhu, but I didn't care for it either. The basic idea of freely giving clues is sound. But the GUMSHOE system was developed as an attempt to correct a fatal flaw with using skill checks in a mystery scenario, namely the danger of the whole investigation screeching to a halt when the detectives fail a crucial skill roll. Basically, GUMSHOE turns "core clues" (ones that further the plot) into giveaways (automatic success) but allows additional information or other benefits to accrue if a high skill roll is made. However, if one does not use skill rolls to begin with, the GUMSHOE workaround is irrelevant. If one does use skill checks, GUMSHOE kinda nerfs the dice themselves. Trail of Cthulhu favors a Ball of Twine model for mystery design (see my post on GURPS Mysteries above) with a series of "scenes" strung along a breadcrumb trail (hence the word Trail right there in the title).

  12. However, if one does not use skill rolls to begin with, the GUMSHOE workaround is irrelevant.

    Yes, that matches my impression too. I think the degree to which one likes GUMSHOE depends greatly on how invested one is in the notion that skill/ability checks are the best way to uncover information in a roleplaying game.

  13. Speaking as someone who loves mysteries but is not good at solving them (except through genre tropes or intuition followed by looking), I think players usually should fail to solve mysteries. The NPC master detective is always there (even if the players think he's the janitor or the dumb cop). You give the players a chance to solve and accuse, but if they get it wrong or get hopelessly stuck, the master detective solves it instead (which would still be fun, especially if he points out some guilty-looking features of PC behavior before exonerating them).

    Of course, if players are really having trouble investigating, you can deploy the NPC detective scenario earlier or just have bitchy but correct suspects or other NPCs hanging around. I mean, obviously everybody involved in a crime except the criminal has reason to hope it gets solved (preferably away from them). It would be a lot better than knowledge rolls.

  14. Personally i like the idea of proper roleplaying for the gm sets a clue up hidden under a loose floorboard. If the players state they check the floor etc...then they auto find it...If not they miss it but hopefully then can look for other avenues.
    Means more work..possibly..for the gm..but also allows the pcs to fail.

  15. Gumshoe is pretty much as you describe this adventure - right place, right time, right question; receive clue. No die roll.