Declaration of intent
Inventing abstract strategy games hardly ever makes it to someone's profession. Public interest is usually modest and among publishers it is even less, so why bother. However, the emergence of the world wide web has made that there are far more inventors now than there were a few decades ago. The numerous new platforms where abstract games can be played, allow them to be presented to the general public more easily and in consequence inventing them has been evolving. I fear that this is for the larger part fuelled by a vision of publishing a game for fortune and fame. That's pretty naive and totally legal. But please don't make it bad games.
For a small minority inventing abstract strategy games is art for art's sake, a quest sometimes bordering on obsession. Insofar as it may be called an 'method', and insofar as fuelling obsessions may be called 'beneficial', they may benefit from 'inside out' inventing. It may render a game in seconds, and the result is usually 'modification proof' with hardly any issues to be resolved. It may also encounter some scepticism by the majority of inventors because of that.


A philosophical question
Here you see a grey diamondshaped hex grid. It has four edges and four corners that are each considered to belong to both adjacent sides. Now suppose you get the idea of connecting opposite sides with a closed path of hexes in one colour, then it is easy to see that this path blocks any path for connecting the other two sides with another colour. Likewise, if you have an incomplete connection, the opponent will always have a possible path for his colour. So if two players alternate in placing one stone of their colour, trying to connect the corresponding sides, one will win. That's a game indeed and it's called Hex.
Now the question is: did Hex 'exist' before someone put down the rules?
If the same question were asked about Chess, then the answer would obviously be 'no'. But what is there to 'invent' about Hex? You see the diamond shape, you see the behaviour - one stone at the time, alternating - and then the rules are inherent. So it's an interesting question.
I'm not interested in the truth of the matter, but in what point of view might benefit an inventor. If you're looking for something that already exists, then you're a hunter rather than an inventor. I'm not sure what inventors do, but hunters look, hunters sniff the air and hunters rely on intuition, and all that won't help you to invent say a new chess variant. There are only a handful of attempts to make a Hex variant, while there are literally thousands of chess variants. At the core of a chess variant is the idea of 'checkmate', but this tells very little about behaviour. Even the great traditionals Chess, Shogi and XiangQi behave very differently. Basically there's an absolute piece that must be captured with other pieces. You can choose their properties and numbers at will. Need pawns? Simply choose them. Cage two like but opposing armies consisting of your chosen pieces on a board and observe their interaction and behaviour. The whole process is pervaded with arbitrariness, but a good craftsman will be able to assemble a proper game.
That being said, I don't feel I can learn anybody all that much regarding this method, whether it concerns chess variants or other games with arbitrarily chosen pieces. These games show only post-invention behaviour. I'm more interested in pre-invention behaviour.

Organic mechanisms
I coined the term 'organic' to describe the behaviour of some games, but I should stress that it is as much a property of these games as it is a way of looking at them. As a property it is vague, but there's an obvious affinity with growth and uniformity of material. Look at a game of Go and a game of Chess in fast forward, then Go looks like a growing organism while Chess looks like a battlefield with pieces killing off their equals from the other camp. As a way of looking at them it is very clear: you simply try to discover aspects of behaviour that look organic. I don't know if keeping an eye on nature helps, but I remember a very pervasive vision where I suddenly saw the behaviour of Go as that of two sets of conflicting bacteria in a Petri dish, each with its own will and intent to grow till forced to grind to a halt. That last aspect is important: can you see an envisioned behaviour actually terminate?

Core behaviour
Sometimes you may suddenly see behaviour that inherently already follows certain rules, turn into a complete game without having explicit knowledge of all rules. The rules follow the vision and what was seen in a split second may take some considerable time to explain, depending on how complex the resulting game turns out to be. Ideally the vision is completely self-explanatory, and if not, then certainly for the larger part. If the emergent rules lead to problems that need regulation you should look deeper: if the core behaviour is sound, then the rules will be there.
I call this 'inside out inventing' because it is based on core behaviour turning into a game. It doesn't require an a priori goal and there's no saying whether it will be a strategy game or a tactical one, or how complex it will be. That all depends on the behaviour that was spotted in the first place, and the game will reveal what it is once it has become what it is.
Alternatively, if you decide to make a specific kind of game you may start with the requirements for such a game as you envision them and you may adapt whatever mechanism fuels it to those requirements. That's a perfectly normal way to go about it, but it is somewhat at odds with 'inside out' inventing because in the latter you're not looking for anything specific. On the other hand it is more of a way of looking at games than it is a 'method', so whatever you do may not matter all that much. I've invented games in unlikely circumstances, none of which, by the way, involved boards or pieces.

It's alive!
Now you're a hunter so you prepare. You know a thing or two about grids and pieces and objects and protocols and the like. But you can't force a vision, so what do you do? In my experience you can't do anything to force a vision, but you can do anything so long as you keep your eyes open. You never quite know how and where you might spot prey. You're looking for something alive, with behaviour that shows will and intent and a goal and finitude. The material is uniform, but the protocols regulating behaviour are varied and manifold and there's no particular lack of goals either. So new combinations of them are endless and in this sea of undiscovered games, many gems may still be hiding waiting for someone to spot them.
Now 'spotting' them is not a method, it simply requires open eyes and an open mind. An open mind is aware of the fact that unrealised unjustified premisses lure everywhere to narrow one's view. Thinking about games helps, fumbling around with boards and pieces doesn't.

Counter examples
Let's start with two of those, for contrast. I remember deliberately inventing some games, for instance for a contest, but they often turned out to lack life and I eventually ditched a lot of them. That says something about 'outside in' inventing, where you have a specific type of game in mind with restrictions regarding goals or mechanisms. It is a process where you try to find a behaviour that matches the means and the goal, not the other way around. Although it runs counter to inside out it inventing it may yet turn out nicely. Here are two examples that both came the hard way and that both behave fairly organically:

Havannah was inspired by Hex. I actually wanted to improve on Hex, but then, it was my first game. So I drew a hexagonal board with a base-8 triangular grid and tried out possible goals that could be reached by single placements. Eventually I got lucky when I saw the 3-fold goal of a ring, a bridge and a fork actually work. I'm not sure if this was due to 'intuition' but it indeed turned out to be a perfect merger. The first games were of course terrible, because seeing it work and playing it well are two different things. I can't stress that enough. But eventually Havannah became what it was: a modest but true strategy game.
But it was pure beginners' luck. I had no vision of any behaviour except alternate placement, because I was still searching for a goal fuelling it.

Inertia was inspired by Ayu and actually started out as an Ayu clone. Till that time I had largely disregarded the goal of 'unification' but the very organic character of Luis' game had nestled itself in my brain. So I went 'outside in': I had a specific goal in mind and started looking for matching behaviour. It became an obstacle race that lasted for weeks because I couldn't quite shake Ayu's mechanism. So it was again sheer luck that I eventually found a neat protocol with a clever restriction that turned it into a nice little tactical game. Despite shrinking material it shows a fairly organic character because its object makes it strive to become 'one'. Quite as organic as LOA actually, but less prone to opening analysis.

Two examples of a 'vision'
For contrast, recently Starweb dropped into my lap and you can read what happened here and here. I was thinking about a system of superadditive scoring called 'group penalty' that basically gives an incentive for groups to connect. It usually goes from the premiss that there are 'value cells' and 'connection cells' and makes connecting groups with value cells advantageous. At the same time the screen showed a Superstar board. Now I didn't conciously think "hey, six inward and twelve outward corners, that's a nice set of value cells", but the notion was there. Then came the vision.
I don't know how 'triangular scoring' entered the equation, but when it did it was as if the vision literally collapsed into a game. Suddenly groups had an incentive to connect, and an even better one than 'group penalty' because the reward for making connections, or breaking them for that matter, would be variable. 'Group behaviour' on hex grids is well known so I could literally see groups striving to connect using maybe slightly different yet very familiar tactics. Triangular scoring fuelled a new core behaviour! Groups were basically starting from 'value cells' at the corners and fighting, connecting, cutting towards the centre, going 'outside in' (if you appreciate the irony). At the periphery of the vision came the notion that 'triangular 9' equals 45, so that counting would hardly be a problem and the notion that inward corners play a different role than outward ones. Of course the behaviour would implicitly end. It was a complete game, instantly.
I immediately noticed two issues requiring regulation, symmetrical play and decisiveness, and wouldn't it be nice (I said in the very first post at BGG) to solve them simultaneously with one rule? And that rule - in case of an equal score the player who placed the second stone wins - presented itself effortlessly. It builds on the almost inherent use of the pie rule: the 'placer' and the 'chooser' both have to take it into account.
As a bonus, Starweb turned out to be a strategy game, very reminiscent of Havannah. The basic strategic dilemma is whether or not to sacrifice a corner to have more influence in the centre. And that one comes early and is very deep.

Where Starweb turned out to be structurally simple, Mu became extremely complex. I've described its emergence sufficiently in About Mu so here I'll try to focus on the sudden vision of 'core behaviour', because after that the rules were largely self-explanatory.
It was around midnight and I was riding home in a state of confusion after an encounter with a monumental but weirdly flawed multiplayer abstract territory game called Atlantis that Martin Medema had introduced that same evening at the games club Fanaat at the University of Twente. What happened requires some conceptual context.


This is a game of Atlantis in progress. The board is made up of 7-cell segments so cells have between 3 and 6 neighbouring cells. There are six players, yellow, green, blue, red, orange and purple. White men indicate 'wells', black men indicate 'craters'. Craters are no longer part of the board and wholly cratered segments may be removed entirely. You can see that happened at the bottom while top-left is a fully cratered segment inside the board that has been left in place to prevent a hole. The general idea of the game is based on:

  • The Focus way of movement, with stacks traveling a distance equalling their height.
  • Cells having a 'capacity' equal to the number of adjacent cells, and 'exploding' the moment that the number of men in a stack equals or surpasses that capacity. The content of the exploding cell is then distributed over those very neighbours, one man to each cell, with any excess evaporating into thin air.
  • Exploding cells turning into into 'wells', growing a new man every turn till the number reaches or surpasses (!) capacity.
  • Exploding wells turning into into 'craters', dead cells that are no longer part of the board.
  • Capture by replacement, regardless of a stack's size.

There's positive feedback involved. An exploding cell not only gives a man to every neighbour, but it also reduces the number of neighbours of every neighbour by one, namely itself. That has 'chain reaction' written all over it. These chain reactions do not bring territory, they kill it. Players must take extreme care not to be eliminated by their own actions. Energy 'evaporates' in the process. Wells cannot explode their content to other wells: they need normal adjacent cells (whether or not occupied) for that. So some wells never sprout because their neighbours have already been blown away. Only the centre provides some refuge - that's why it is so big. There your wells can grow and sprout, and your craters become 'walls'. But its extremely hard to plan an efficient wall using explosions that may go out of control, let alone achieve one.

Atlantis is a conceptual disaster, that's what I thought on the way back. Building walls with explosions! Ask any person what explosions are used for, to build something or to clear something, then the answer is clear. And then the vision hit.
I saw, peddling in the dark, very much the same board as in the picture above, but covered with a top-layer. A top-layer that could be cleared away by explosions! Adding territory instead of killing it! Conceptual clarity!
This was the primal vision and the behaviour, goal and finitude followed suit: you begin with one segment filled with your men and the first explosion clears away the top layer of the cell. Any remainder stays on that cell, now a bottom-layer cell, and the cell sprouts one new man. Thus growth is inherent, but modest. Now you can see the following scenario, an exploding cell of the top-layer will

  • Expose a new isolated cell of the bottom layer
  • Expand precisely one area of the bottom layer
  • Threaten to expand more than one area of the bottom layer

If an exploding top layer cell threatens to merge two or more bottom layer areas, then the explosion turns it into a cell of 'the Wall'. That means that seperate areas of the bottom layer never merge. Once the whole top layer has eploded away, it leaves a web of walls that together constitute 'the Wall', separating the territories between them. Beautiful!
It was clear to me, still peddling, that the basic behaviour should continue on the bottom layer. Top-layer cells would inherently only have top-layer cells as 'neighbours' to keep capacity low and speed up clearing. But bottom-layer cells should have both top- and bottom-layer cells as neighbours. Think about it.
Then, still peddling, it occured to me that territories, by the nature of the beast, would hold stacks of differnt players, fighting it out till one would remain. That's ok. It also occured to me that if an exploding top-layer cell turns into a cell of the wall, it might do so with 'overcapacity' - what to do with that? If it happens to a normal bottom-layer cell, then the remainder stays where it is, so why not keep it that way. It creates the possibility to have men on the Wall - what can you do with them? Well, you can use them to invade an adjacent territory, obviously, but maybe also as a bridge? To allow men of that colour to jump the Wall over such an occupied cell?
That's where I was when I also suddenly found myself peddling close to home. I was elated! Of course where Starweb had two remaining issues, Mu must have had at least twenty. But their solutions were easily guided by the intention of the beast itself. You're dealing with a living organism, remember? It wanted freedom and growth and it showed the way by extending strategic and tactical options in the process.

I invented Mu for one reason only: because it was there when I saw it. It turned out unusually complex for an abstract strategy game, but it did so from a core behaviour that shows unity, and that unity was maintained, even deepened, in the game's becoming what it is. I can only recommend it (you can play the two player version at mindsports if your browser supports java) and I hope the story of its birth helps to understand how the rules, elaborate as they may be, support a single purposeful organism.

Four of the 'five that matter'
In 'Moving forward looking back' I mention five games that I feel do matter, Grand Chess, Dameo, Emergo, Symple and Sygo. No less than four of them were invented 'inside out', that is: starting from a core behaviour in which I saw will, intent and finitude.

In Dameo's case I saw how linear movement could bring flexibility of pace to a Draughts game. I know International Draughts very well, so I naturally 'saw' it in action in that game. And I saw a good chance of gridlock in the opening. The vision was so clear that I didn't even try. Nor did I consider a possible cause, I just put the idea on the shelf. There it remained for fifteen years and along comes Croda. Magnificent game! Then, after a couple of weeks the inevitable happens: I suddenly see linear movement in action in Croda. No gridlock! And Dameo emerged in two minutes.

I saw Emergo the very moment Ed van Zon showed me 'some beautiful things' that he had seen happen in Stapeldammen. They bypassed the checkers jacket and showed how the organism yet terminates by its own mechanics! I saw the core behaviour and it was poetry in motion. A beautiful organism that had been jammed into a checkers cage! So we freed it, there and then, within an hour.

Symple emerged as a result of the clearest example of a vision, because it happened while sinking into sleep, and it was only late in the next afternoon that I remembered it and could mail it to Benedikt Rosenau, well, sort of. The vision was not so much a game, but a move protocol: you either create a new group or you grow any or all existing groups of your colour. I mailed it in a hurry and actually implied that both actions could be made simultaneously, but Benedikt immediately noticed that it should be one or the other. It creates a dilemma in terms of growth: to grow fast you need many groups, but creating new groups prevents you from growing the others.
As an unexpected bonus, the protocol turned out to have an embedded high resolution turn order balancing mechanism that is totally fair.
The protocol is generic and though Symple came first, it carried a bug for a couple of months because I went from the premiss of an option to pass, as in Go. That was wrong and I knew it because the game seemed to lack drama. But I needed the clear vision of Luis BolaƱos Mures to have it pointed out and correct it: moving became compulsory and a growing move must include every possible group. Symple isn't Go, but a territory/connection hybrid with its own rules. But in the meantime the protocol had sparked Sygo, which is a true Go variant featuring optional growth and the option to pass, and terminating after successive passes. Sygo is based on Othelloanian capture, aka 'flip capture' to prevent cycles, and it is so far the only such variant that doesn't need an additional criterion to ensure life. Its organic behaviour creates its own inherent means to do so.

Conclusion
I tried to clarify that 'inside out inventing' depends on seeing purposeful core behaviour. Such behaviour inherently suggests a specific goal and should inherently have finitude. The problem from an inventor's point of view is that it is a way of looking at games, rather than a 'method'. Usually you start with a specific goal in mind. I wouldn't discourage that because it's almost inevitable. But it's like trying to blow life into a mechanism and it has a considerable failure rate. 'Inside out' is based on a vision of an almost living thing. From such a vision a game will reveal itself in a largely self-explanatory way, and often almost instantly. And it won't let you end up in Tinkertown, modifying till pigs fly.


christian freeling
november 2017