Voting Decisions

„We are now in a position to present a model which attempts to integrate the various models in a fashion which incorporates the features just discussed. That model is displayed in Figure 1. The first two steps are the same as the first two in the consensus mode of decision, which I have presented elsewhere. If there is no controversy in the environment at all, the congressman’s choice is simple: he votes with that environment and is done with it. On many bills, for instance, a unified committee reports the bill and nobody opposes the committee position in any particular. If there is some controversy, he subsets the environment, considering only the actors which are most critical to him-his own constituency, his party leadership, his trusted associates in the House, his own policy attitude, etc. – which I call the „field of forces“ which bear on his decision. If there is no conflict among those actors, he votes with his field. I assume, as a legislator does, that if there appears to be no consideration which would prompt him to vote in a way different from that toward which he is impelled by every factor in his field of vision, then there is no reason to think twice. And as I have argued above, this is a beginning to an integrative model which is common to a number of the previous works on legislative voting.

If there is some conflict among the congressman’s relevant actors, he then proceeds to consider his goals, which I conceive for the purposes of this paper as being the three discussed above-constituency, intra-Washington influence, and public policy. But a goal is not brought to bear on the decision if it seems unimportant to him on this issue. It must pass what I have labelled a critical threshold of importance in order to be evoked and relevant to the decision. For example, a congressman’s constituency may have a vague and largely unarticulated opposition to foreign aid. In that case, he would say that there was a constituency opinion on the issue, but that it was not intense enough to bother taking account of. The same could apply to the other goals. In the next section of this paper, I present some operationalizations of these thresholds and use them to deal with data on voting decisions. If none of the goals is important enough to the congressman in the given decision to be relevant, he then proceeds to follow trusted colleagues within the House. He chooses colleagues who are on the committee that considered the bill and who agree with him in general philosophical, policy terms.22 If one or more goals are important enough, he asks if there is conflict among the goals which have been evoked. If there is none, the choice is then clear: to vote with the evoked goal or goals (Step Cl). It could be in this case that only one of them is relevant to the decision, or that two or even all three are, but that they all point him in the same direction. For example, it could be that the policy goal on a given issue is the only one which passes its critical threshold, and the other two, while either opposed to, favorable to, or neutral concerning his conception of good public policy, are not in any event important enough to him on that issue to be potentially controlling. He votes in that case according to his conception of good public policy. As the model specifies, part of this decision may well be picking cues within the House to reinforce his policy goal, as a means to that end, in the fashion discussed above. Other examples of no conflict among evoked goals could be given, but this one will perhaps suffice.

If there is some conflict among the goals which the legislator considers relevant to his decision, he proceeds implicitly to some decision rules which help him sort out the conflicts and make a satisfactory choice. It might be helpful at this point in the argument to present all the logically possible combinations of conflict among the three goals, which is done in Table 1. In the first column, the possible combinations are listed, and the second and third columns contain the outcomes which the model would predict for each of the combinations. The numbers are relevant to the operationalization, which is explained in the next section of this paper.