In the meantime, here we have one way in which reasonings about fact do help to modify our actions. If we know distinctly what we desire, say, a week’s holiday, then it is reasoning about matters of fact that will tell us what we must do to get it, and, in part, what the results will be both of our getting it and of what we do to get it. Now for philosophical purposes we need not consider the foreseen consequences separately. They must rank, morally, as means. That is to say, they are something which you have to take into the bargain in order to get what you want. They come in with all the other circumstances in determining whether you like the action or not. Now is _this_ connection between action and reason what we have in mind when we say that a person ought to act reasonably, or that reason combats desire? Do we understand by acting reasonably, that assuming some one part of the imagined circumstances to represent the purpose, the agent has got all the means to it, and the foreseen consequences of it, and the interdependence of the parts of the purpose itself, set out in a connection which is truly perceived or scientifically inferred? We do sometimes appear to mean this. We say: it is unreasonable to ask me to be at the station at nine when the train does not start till ten. It is unreasonable, you may say, on the ground that the means demanded of me are not, scientifically speaking, necessary to the end agreed upon. Still more we should pronounce it unreasonable to adopt any means which actually defeats your purpose; which could only happen, one would think, either from moral self-deception, or in complicated matters where the means are disputable. This second case does not matter to us; the first carries us a little further, because it suggests that what you call the means may really contain your purpose, or one of your purposes, perhaps inconsistent with another. The hackneyed example of selfish charity is as good a case of this unreasonableness as can be found. The gift, which is professed to be merely a means to the good of another, is, under all the conditions, a means contrary to that good, and is given because it gratifies an impulse of the donor. It might seem, in this case again, a fair explanation to pronounce such conduct unreasonable merely because the means adopted are scientifically speaking inconsistent with the end proposed. We might bear in mind, however, that we seem to have detected here a probable conflict of ends, not merely of means to an end.