Section 9 Metalanguage: some words for talking about causal maps

We will need some metalanguage to talk about causal maps. Metalanguage is not part of the maps, it is a part of English which we use to talk about the maps. I’ve spent quite a lot of time trying to get the terminology right, I’m not making up words for the sake of it.

9.1 “The cause”, “a cause”, “(causal) influence”

I prefer not to use phrases like “the cause” or even “a cause”. For one thing, these are too mixed up with our human concerns like blame, and legal responsibility, and moral judgements. For another, these words are too specialised for binary, true/false variables. Also, they are too monolithic. For example, C is certainly not “the cause” of E below.

For similar reasons, in the diagram below I prefer to say that C influences or causally influences E. Indeed we can and should also say that the package of B and C together causally influences E. This makes it clear that we are talking about causation, not for example correlation, but that we are not of course claiming that E is completely determined by C or B.

9.2 Nodes, variables, vertices …

From a network perspective, the boxes can be referred to as nodes or vertexes. I will refer to these mostly as variables. However, calling them “variables” means that they are things which can vary, be different, for example the temperature outside now, or the extent of last season’s rains; sometimes, variables don’t seem like quite the right fit when coding causal information which talks about events. See later.

The lines are referred to as (directed) edges, paths or links. I will call them arrows.

9.3 Influence variable, consequence variable, package of variables

In a map like this, we call the variables on the left the influence variables of E, and E is the consequence variable of B and C. We can also say that B is a parent of E and E is a child of B. And we can say that both E and F are downstream of B and C, and that they are descendants of B and C, and so on. We can also say that B, C etc are members of a causal package – provided they are indeed coded together as such, see xx.

9.4 Mechanism, theory ….

It can be useful to distinguish between on the one hand “theories”, “theories of change”, “causal maps”, “representations”etc on the one hand and the aspects of the world which these things are true of on the other hand, which we can usually refer to as “mechanisms”.