Simplicity is on the side of separation. Linkage, on the other hand, demands complexity. The first step in establishing a link is categorizing all skills. It seems the easiest possible way to do this is to decide which attribute governs each skill, but therein lies the thorniest of problems. How can one say that Agility alone governs a skill in Running? What about Strength? What about Endurance? Who is to say that Intelligence alone governs a skill in Drawing? Perhaps the creative aspect of Spirit or Soul is involved. Perhaps the skill is largely owed to Perceptiveness or Manual Dexterity? In response one may argue that the solution is to create new skill categories composed of combinations of skills, as in Basic Role-Playing. Categories such as Fine Manipulation combine Intelligence and Dexterity. Unless each attribute is used the same number of times, however, one or two attributes will dominate the skill categories (usually Dexterity and Intelligence or their equivalents), and players will min-max with wild abandon. It’s not so much that most players are power gamers, but that the temptation to take advantage of the system is just too great. The mere presence of a linkage not only enables min-maxing, but actively encourages it. (And this, incidentally, leads to the original and most compelling argument in favor of separation as made by the author of Fudge, Steffan O’Sullivan, in his Recent Thoughts comments).

The next step in establishing a link would be deciding how the attribute and skill actually interact. Do attributes confer a bonus or penalty to purchasing skill levels or as a modifier in action resolution? Do unskilled actions default to a governing attribute or skill category? The problem with this approach is skill inflation. Either no skill can be taken at face value (which defeats one of the basic design principles of Fudge) because it may be raised or lowered by an attribute, or no unskilled action can be counted on to be justifiably inferior to an action by someone possessing the skill at a moderate level. In any event, what you see is not necessarily what you get and thus another layer of murkiness complicates the game.

In the real world, all of the factors involved in connecting whatever attributes we possess with whatever skills we have learned are complicated beyond any enumeration or comprehension. Instead of trying in vain to simulate something so complex, why not use common sense? If a character has many academic skills, why not assign an appropriate level in the appropriate attribute? In terms of skill acquisition, why should a character who is a mathematical genius receive a bonus to acquiring skills in Writing or History for example? One kind of mental excellence does not necessarily translate to another, and the same goes for any other attribute. The attributes used by a hockey player do not necessarily translate into the same advantage for learning flamenco dancing just because hockey and flamenco are both physical in nature.

The Fudge solution is elegant. I resisted it upon my first reading of Fudge years ago, but I came to realize that it solved one of the worst banes of game design, and it did so by relying on a common sense approach. For that reason, I will maintain the separation in my own games. I may design or run games that have attributes and no skills, skills and no attributes, plain traits, or separated attributes and skills, but I’ll never again resort to the legacy of linkage when it comes to Fudge.