additive

respectively, where A = additive relationship matrix; I = identity matrix; [[sigma].sup.2.sub.fi] = farm variance, [[sigma].sub.f1f2]= farm covariance, [[sigma].sup.2.sub.ai] = additive genetic variance, [[sigma].sub.a1a2] additive genetic covariance, [[sigma].sup.2.sub.ei] = residual variance, and [[sigma].sub.e1e2] = residual covariance.

This means that (i) there need not be an explicit question, as in the examples above, and (ii) the utterances linked by this additive relationship can be nonadjacent.

While it is certainly true there is a (in my view, unfortunate) tendency within much of the causal modeling literature to focus almost exclusively on additive (and, indeed, linear) models, this seems to derive mainly from considerations having to do with computational tractability and from a preference for certain simple statistical techniques and perhaps also from a desire to impose an artificial limit on the space of alternative models among which one needs to search rather than from any assumption that additive relationships are especially suitable for expressing causal connections.(4)