Monday, January 21, 2013

Strategos in the Twin Cities

Twin Cities gamers of the late 1960s found inspiration for their wargaming systems in a pretty unlikely source: Strategos: The American Art of War (1880), a work by Charles A. L. Totten. Strategos was a military training wargame modeled on the German precedents of the nineteenth century, and was therefore not designed with entertainment in mind. By modern standards, it is unwieldy, pedantic and dull.  Famously, Dave Wesely adapted the rules of Strategos to the Napoleonic era, reducing them from the two volumes of Totten down to just twenty-four pages in his Strategos N (1970). But the Napoleonic era was only one of the settings that local gamers fitted to Strategos, and Dave Arneson himself authored two Strategos-based wargames.

The most important element that Wesely extracted from Strategos was the impartial referee who oversees a wargame and determines the results of players' actions: as I argue in Playing at the World, the concept of a referee is essential to enabling a game where "anything can be attempted," where players can propose arbitrary actions as they do in modern role-playing games. From Strategos, the idea of the referee entered Blackmoor, and then ultimately Dungeons & Dragons. Beyond this high-level playstyle, Strategos N also incorporated and simplified the combat system of Totten, most notably in its adaptation of "Table T," the standard combat resolution table of Strategos:
Wesely was working on an adaptation of Totten early in 1968, and by 1969 numerous local gamers were experimenting with the system; his stripped-down "Table T" became a staple of Twin Cities gaming. In January of that year, Arneson himself was already using Strategos as the basis for a Russo-Turkish war campaign he was running. These rules are preserved in his Strategos RT (1969), which circulated in a draft form along with Wesely's early Civil War Strategos adaptation, Strategos C (which we should probably understand as the first step towards his Valley Forge (1976)). Later in the year, Arneson circulated a version of Strategos adapted to the ancient setting, Strategos A (1969), which he wrote along with Randy Hoffa.
Today, most of these Strategos variants are long forgotten. Since Totten is in the public domain, the original 1880 Strategos is widely available online and in print form, though its second volume, which contains the tables, seems to be more scarce. Wesely did attempt to make his Strategos N rules available to the world one more time, in 1984, through a new spiral-bound edition with a cover by Ken Fletcher. Perhaps the best feature of this second edition is the contemporary image on the back of Wesely himself dressed up as the commander of the Grand Armee. Practically speaking, however, this edition is now as rare as the 1970 original.


  1. I see your point, but I can't help but think that it should not have taken a substantial amount of inspiration to decide that a referee would be necessary. At a certain point, a game just becomes so complex that it almost becomes intuitive to determine that a referee is necessary. Perhaps the big innovation with respect to roleplaying was not the addition of the referee but the elimination of a live opponent.

    Also, I am inclined to speculate that referees would be involved in games that made no mention of the necessity of a referee. In those games, a referee would take the place of a more experienced person teaching and mediating between two novices.

    My own intuitions make me inclined to think that when a game author requires a referee as a part of the game experience, it is because the author knows that a circumstance not covered by the rules is likely to occur. I have not read the Totten game or any of the later adaptations; my conclusions are only what seems reasonable to me.

    IMHO, what makes the referee necessary for Dungeons and Dragons is the inclusion of unaligned persons and creatures in the game setting. Where the previous need for a referee was the inability to account for all of the actions that a player may attempt, now the rules have been refined and expanded so that a rule can exist for every imaginable thing. Many may be arbitrary, and many may be at odds with reality, but the players can acknowledge the rules and move on with the game. What makes a referee continue to be indispensable in D&D and other roleplaying games is the need to determine the actions of the individuals or entities in the game who are neither the players nor the players' opponents.

    I think you make a good argument that the existence of a referee as a necessary participant in in the Totten Strategos game is the proximate cause of the appearance of a referee as a necessary participant in the Wesley and Arneson Strategos variants. Consequently, I think you also make a good argument that the appearance of a referee as a necessary participant in Wesley and Arneson Strategos variants is the proximate cause of the appearance of the dungeon master in the original D&D game. However, I think that the range of actions can only partially account for the necessity of the DM.

    Of course, it may have been the complexity of the game that was the motivation for the inclusion of the dungeon master--I have no way of knowing. The development of nonaligned entities may have been an afterthought in the minds of Gygax and Arneson and company. I personally find that the treatment of nonaligned individuals is greatly overlooked in videogames and MMOs.

    1. The full argument given about this in PatW would be hard to summarize in this space, but as a question about history I think this is relatively straightforward. Of course there had long been referees to adjudicate wargames, and those referees did not by their very presence cause role-playing to occur. However, what Wesely found in Totten (which Totten ultimately owes to Reiswitz) was a referee with the specific purpose of interpreting the verbal orders of players as the sole way that players may "move" in the wargame. Typical wargames had far more rigid systems with far less freedom of agency - while referees may have been a tacit presence, especially at tournaments, the moves in these games did not require a referee to interpret them. Once you introduce the ability of players to say "I try to do this," where "this" could be pretty much anything, you are well on the way to role-playing as OD&D cast it.

      The need for a referee to intrepret moves is in part based on the presence of secret information in games, and that is perhaps closer to what you mean when you talk about game entities not under any player's control. Dungeon maps are probably the best example of this secret information in OD&D, but in wargames as well (from Totten, and again back to Reiswitz) the secret information retained by the referee let the player experience the same ignorance of conditions that they would experience on the battlefield as a real commander. Moreover, the player controlling forces interacted with them only through written orders, again interpreted by the referee, so you had to be very careful with phrasing - like D&D referees can make players very careful about phrasing wishes. All of these factors create an immersive experience for the player, make them feel more like the game situation is real. That is another key enabler for role-playing, and another sense in which the referee is needed to create the right conditions for role-playing.

      It is hard to fix any exact historical point at which role-playing became a fixture in these wargames, but certainly Braunstein and Blackmoor had it, and just as certainly Braunstein and Blackmoor did not eliminate live opponents. It is often overlooked that Blackmoor was a team-based wargame campaign that pitted good guys against bad guys, yet role-playing flourished there. It also flourished in Tony Bath's chaotic Southampton games, where were every-man-for-himself wargame sessions (I write about them in PatW 4.5.2). Finally, Diplomacy was a major catalyst for role-playing, and it too was a game full of opponents and coalition-building. So in short, the historical evidence shows that role-playing emerged in games that did not eliminate a live opponent, even though most things we think of as RPGs today do not have that oppositional structure.

      If you're interested more in "nonaligned individuals," PatW can moreover point you to earlier "one-sided" wargames in whch a referee ran a single player through wargame scenarios as an educational exercise, an idea that had been around for most of the 20th century. I certainly note in 1.10 that Arneson's demonstration of Blackmoor to Gygax in Lake Geneva necessarily downplayed the multiplayer aspects of Blackmoor, since the cast of local players didn't make the trek with Arneson. However, role-playing as we know it had been going on in the Twin Cities for some time before then.

    2. Upon reading your response and re-reading my own statements above, it has become clear to me that I was unclear as to what I was talking about when I referred to "other entities" and "nonaligned individuals." Within the context of a roleplaying game, I would be referring to nonplayer characters whose role in the game would be nonadversarial. I am only speculating as I have not read any of the wargames that predate Dungeons & Dragons, but I am assuming that there is no analogue to a nonadversarial nonplayer character in the older wargames.

      I think you make the more general and therefore better point when you discuss unknown information. This did not occur to me since I'm used to having lots of random information provided in games through the means of random tables to be consulted after throwing a die or dice or even a stack of shuffled cards. It is conceivable that Totten could have done away with the referee and instead substituted a device for the random generation of event or circumstance unknown to the players. However, it appears that he did not. I'm guessing it is impossible to know whether he considered random generators in place of the referees or whether the thought never crossed his mind.

      Curiously enough, I have tons of AD&D supplements that provide me with countless tables for generating events. Almost anything you want can be found on a table if you know where to look. Do you need a random wandering monster for your party to fight? Do you need to know what kind and how much treasure the party found after slaying the monster? What about the temperature and the weather? Need a new NPC for the party to encounter? And my favorite: What is the next rumor that will come out of that NPC's mouth?

    3. In the context of wargames, it seems that a referee is expedient rather than necessary. Of course, the referee might have been perceived as necessary by Totten et al. In fact, that most likely was the case from what I've read here and in your book (which I have not fully read--only halfway through chapter three). So really, I have no issues whatsoever with your historical analysis. What I'm probably doing is second-guessing Totten's perceived need for a referee (and over a hundred years after the publication of his game at that!) But I can't really say that I would not have done the same, because it appears from what I have read (i.e. the secondary sources you have authored) that the use of a referee is clearly a simpler mechanism than the use of tables and other random generators.

      Then again, random generators clearly appear as can be seen in your visual aids presented above and in 3.1. So there are some circumstances where unknown information can or should be determined randomly and some circumstances where the information can or should be supplied by the referee.

      I think that a referee, while often expedient, can be obviated with enough preparation. However, a dungeon master remains necessary no matter how big one's book of tables. Your own example with regard to a dungeon master's scrutiny of the wording of a player's wish illustrates this matter perfectly.

      I agree with your statement that Diplomacy was a major catalyst. However, I have never played that game, and I cannot remember from my reading of Chapter 1 whether a referee was required in Diplomacy. It seems to me that it would be much more difficult to play a game of diplomacy than it would a wargame. You can make a table to determine the outcome of any action if it's physical in nature. However, if the action taken is not physical--if it is nothing more than the employment of language--how can you quantify the chances of success or even catalog every potential variation on the action? It is when communicative actions such as parlay, diplomatic gestures, attempts to persuade, and the like are employed that the value of the referee is truly realized. (Of course, it is just now as I write this that I realize that many of the rulebooks provide a shortcut around matters like these by allowing the character to roll a "check" using one's statistics as a modifier)

    4. Going from a game that only involved adversaries to one with a referee is definitely a necessary factor for the roleplaying game to appear. Removing layers of abstraction would also be a necessary factor. Also necessary is the ability to engage in more than just physical acts. It's the ability to employ verbal acts--acts that cannot be reduced to physical action or physical movement. Based only on what I have read thus far (and a limited memory of it at that)I'm inclined that the greatest creative progress came in the form of the diplomacy games played. I advance that only as my opinion, however. And not a well-founded one at that.

      But that's the factor that gives pen and paper roleplaying games their greatest shot at remaining relevant. Non-RPGs just don't consider the existence of characters that aren't necessarily adversarial to one side or the other. And to my knowledge, non-player characters remain hopelessly wooden and underdeveloped in MMOs and other computer/console games.

      I think you have identified beyond controversy when a referee started to be used. However, I am curious as to where the dynamics of the game became such that the referee became indispensable. I would have guessed Diplomacy just from the name of the game, but my first reading of chapter one did not leave me with that impression. Also, I have been in multiplayer coalition-building games (mostly multiplayer Magic TCG games) that can transpire without a referee. Could it have happened in Braunstein? Blackmoor? Was it not until the first Greyhawk game was started?

      Perhaps my line of inquiry is too subjective, or it is more of a qualitative inquiry that eludes precise treatment. And of course the intrigues that go along with the game's early history appeal to me as much as the Books of Kings in the Hebrew Scriptures. However, the question arose in my mind. Of course that not to say that the question is one that is necessarily capable of being answered. I am not, at this point, certain that it makes sense. But since you had time to spend giving my last response a serious answer....

    5. Perhaps I can restate what's on my mind this way:

      A game can begin with certain unknown or secret elements, each of which will be determined at some point while the game transpires. Some can be determined through an engine for determining outcomes randomly. My own perspective is that during the course of a roleplaying game, there are some decisions that can be made with a random decision generator and some decisions that can only be made by a human referee. I believe that although human referees can be expedient for wargames, but they really only serve the purpose of (1) revealing a circumstance that was unknown to a game or (2) settling a dispute not properly covered (or covered at all) by the rules. Since (1) above can be settled by the incorporation of a table or set of cards, etc. into the game, and (2) can usually be settled by a coin toss, I conclude that an element exists in most roleplaying games that do not exist in wargames. I also believe that this element is absent from computer games and will perhaps continue to be absent until a game presents with the ability to pass the Turing test.

      Then again, maybe the element that I have constructed in my mind is nothing more than a figment of my imagination and this quintessential element of the roleplaying game is nothing more than the will of a cruel and capricious gamemaster. And since the grandfather of roleplaying is also the author of The Tomb of Horrors....

    6. The system of Reiswitz's game, all the way back in 1824, made the referee indispensable. Before Reiswitz, wargames were like chess, or like Avalon Hill games today: players moved their own pieces on a board when it was their turn, and supervised their opponents' movement when it wasn't their turn. In Reiswitz's game, players don't interact with pieces. They don't see a board except in so far as the referee may show them one to illustrate state (the way a DM might draw a room for you when D&D combat breaks out). Each turn, the referee represents some state of the game world to the players, and the player responds by describing what they do. The referee only imparts information about the game world that the player would know, in their circumstances. It is simply not possible to structure a game that way without a referee; the referee is then already indispensable. Totten inherits these techniques, and from there so does Braunstein etc.

      It may also help to remember that the need for secret information is motivated in part by keeping unit movements confidential until the forces in battle discover one another. Imparting only such information as a player might know means hiding from the player any knowledge of opponents' positions and forces until scout reports return, for example. Practically speaking, this means a player may never interact with his opponent - all play is mediated by the referee, and it would be safe to say that nothing would change from a player's perspective if in fact there is no opponent and the referee is controlling the opposing forces. The same is true of any allied forces on the map. This all, again, serves the objective of creating a compelling simulation of command, to better train the wargamer for real-life strategy and tactics. It just so happens that these techniques also work well in the context of entertainment.

      If you're on Ch 3 of PatW, then you haven't gotten to the Diplomacy section yet - it's in the beginning of Ch 4. Broadly, Diplomacy does have some sensitive information, and in its postal incarnation especially, a gamesmaster is required. Many other postal games described in Ch 4 inherit this quality from Diplomacy, and we can show how they directly influenced D&D.

      There is a ton more to say about the interaction between dice, verbal orders and secret information, most of which I cover in PatW Section 3.2.1. They are key components of OD&D, and they all have a particular historical pedigree. On dice, and generating information randomly instead of keeping it secret, do note that there was an early "Solo D&D" system published in the Strategic Review that randomly generated an entire dungeon as you went along as a single-player game (I discuss it in 5.5). It's not likely that the dungeon in question will be very interesting, but, it is a proof that you can use randomness as a (poor) substitute for a referee.

      Not all multiplayer games require referees, of course. M:tG doesn't require them any more than Hearts or Bridge do. In card games, players can manage their own secret information (their hand of cards) under mutual oversight. There are multiplayer board games as well that don't require referees. But in M:tG, your scope of agency as a player is very limited. On your turn, you drop a land, tap some lands, play some creatures or spells, and so on. You can't on your turn decide to say "actually I'm going to chop down this Forest and make an enormous wicker statue of myself." All you can do is play cards. That's why you don't need a referee in M:tG. In D&D, however, you can propose to the DM that you make a wicker man, and the DM can decide whether or not that is feasible (perhaps consulting charts or rolling dice as a part of making that decision). The critical thing to understand here is that wargames that came out a century or more before D&D already had systems based on this principle.

    7. Eureka! I had to re-read 3.1.3, but I now see why the referee is necessary in the the younger Reiswitz's game. If I might quote from page 230:

      "In addition to establishing the general idea and the composition of the opposing forces, the umpire serves as an intermediary for virtually all actions in the game: all movements, all communications and all attacks channel through the umpire, in writing."

      I disregarded at my first reading (1) the umpire's complete control over the setting and (2) the secret means by which players communicated moves. These two items alone make the referee a practical necessity regardless of whether he is necessary in principle.

      I would agree that the role of a referee matches that of the role of a gamemaster in a roleplaying game. Furthermore, the duties and privileges of a kriegspiel referee or a gamemaster appear to be very different from a referee for an athletic event. An American Football referee has no control over the dimensions of the field, and a boxing referee has no control over the length of a round. The Kriegspiel referee, on the other hand, sets the parameters of the game (to include, if I understand correctly, the conditions for success). I would consider that an innovation over the previous established games of which I am aware.

  2. Probably the most interesting part of this blog entry and of the discussions of wargaming in your book are the parts of the discussions pertaining to distribution of the wargames. It appears to be safe to say that print runs were low and distribution was limited to certain areas. I can remember being introduced to wargamers as early as 1978, when I was around five years old. My dad frequented both a camera store and a bicycle shop with owners and/or employees who had an interest in the hobby. I never saw the rules from which they played, but I remember that the gamers had miniatures representing medieval, civil war, and napoleonic eras. I do not recall seeing any rulebooks. However, I am inclined to say that large scale games would be played at a 1:1 figure to man ratio. This strikes me as odd given what I'm learning in the book since the man-to-figure ratios in the publications mentioned above are 10:1 or 12:1. Also, it is my understanding that ratios in Chainmail were 20:1 or 10:1, depending on the size of the figure.

    My initial impression is that these rules were used in my area but modified with house rules for a 1:1 representation. An alternative is that the earliest players in my area played by another set of rules that employed a 1:1 ratio.

    Do you know if anyone has kept records of wholesale/retail distribution? I would be interested to know if anyone in my area purchased anything from Wesley--or perhaps Don Lowry...? If I recall, the gaming was active in my area, and the advent of the roleplaying fad didn't take away from the numbers of wargaming fanatics. I never played with those scary old guys, but they fascinated the hell out of me, and I would love to know what it was they were playing.

    1. There were certainly many 1:1 figure scale wargame rules in circulation in the 1960s and 1970s - PatW 3.2.4 covers this at some length. Chainmail had 10:1 and 20:1 rules, but also 1:1 (man-to-man) rules.

      I doubt that Wesely's game was distributed in any meaningful way beyond the Twin Cities. Lowry's Guidon games, however, sold in pretty substantial print runs. I doubt however that much by way of retail distribution records down to that level survives at this point. It is easier to just look at wargaming zines and find submissions from local people to figure out what they were playing at any given time. Depending on where you grew up, it might actually be easy to find out what was popular in 1978.

  3. Miniatures wargames that require a referee are still being written today. A referee allows the game to take on a much more realistic approach since you don't have to codify all possible player actions. For example, when Heinz Guderian was leading a panzer group during the invasion of France in 1940, he issued the exact same orders that he used for the pre-invasion wargames the German had run. He just changed the dates!

    The problem with this type of "do anything" games is the time it takes for the referee to process all the actions of all the players. Games of this type would take multiple days to run (or were play by mail). The first Braunstien game only got through two turns IIRC.

    The big change in these types of games happened when Dave decided to take the Blackmoor bunch into a dungeon. The confines of the dungeon limits the player's available actions and allows a referee to run a "do anything" type of game in real time. No writing down instructions or waiting for turns to be processed. This interactive nature is what separates an RPG from a wargame with role-playing added on.

    The dungeon also changed the focus of the game from player-vs-player to players-vs-environment and the dungeon delve allowed the players themselves to set their own victory conditions.

    1. I definitely agree that the confines of the dungeon simplify choices and thus make it much easier for a referee to handle requests to "do anything." I'm not sure I follow how that lets the players set their own victory conditions, and the victory conditions of a dungeon seem built-in to the environment, but we're at least on the same page here.

      That much said, it seems to me that once people understood what role-playing was, taking it out of the dungeon wasn't that big a deal. You can role-playing in the Wilderness of D&D, or in a town, just as much as in a dungeon. Many subsequent RPGs don't have dungeons at all. Referees however are still able to process requests from players to "do anything" even outside of that constrained environment. So while the dungeon may have helped point us in the right direction, they certainly are not a necessary condition for real-time role-playing.

  4. Very interesting! Would love to hear more about Dave Arneson's Strategos Games! :)


  5. BTW, when I was at Coleco post-TSR, I worked with a game designer named Ken Totten, a descendant of the Charles Totten of Strategos. He was brilliant at the sort of simple yet elegant abstract games that are so difficult to design, let alone perfect. He was also a really nice guy, and I've always been sorry I lost track of him in the mid-'80s. Wonder where he is today?

  6. Several years back I prepared presentations and article on both Strategos and Livermore's American Kriegspiel. Re the latter, I found that many libraries actually separated the second volume of American Kriegspiel from the first because the second volume was often labeled as an atlas - so the librarians dutifully shelved it accordingly - and separately from the first volume with the text. This may have happened with Totten's work as well.