Triangulation Again, with a Big Twist

If this article in The Hill is true, and it is eminently reasonable, we are in for some pretty interesting politics ahead that will make every spatial modeler’s head explode.

The gist of the article is that the meeting between “very liberal” Hawaii congresswoman Tulsi Gabbard and Donald Trump, which has a lot of people befuddled, was in fact orchestrated from the Trump camp, at the initiative of Steve Bannon, formerly the campaign CEO for Trump.  This comes in the heels of statements by Sanders and other “very liberal” Democrats in Congress where they expressed a willingness in working with Trump where their interests intersect.  It seems that, in effect, the Trump camp is responding favorably to these overtures.

None of these should really surprise anyone:  this was, after all, the core of the Clintonian triangulation playbook.  The idea was, in the end, simply that the White House should be willing to work with factions in Congress that go beyond the party labels and piece together the coalitions behind the presidential programs whichever way they find feasible. At the same time, of course, this is a very big departure from the Clintonian triangulation, which was rooted on the assumption of spatial politics.  It assumed that the “left” and the “right” existed and the “middle” ground between them provided a natural winning position, in the manner of the Median Voting Theorem.  Here, Trump and, more importantly, Bannon, are supposed to be representatives of the extreme right.  Sanders and Gabbard are supposedly extreme left.  So what gives?

The short answer, I think, is that spatial models and its associated notions of “far left,” “far right,” and “moderate” are all bunk.  The longer answer is that what we think are “far left” and “far right” (and the alleged measurements, via DW-Nominate and related techniques) are all figments of our limited imaginations about politics and coalitions.  DW-Nominate and associated techniques derive supposed spatial ideology of legislators based on their voting patterns.  Thus, “far” left means that the legislators voted against “mainstream” Democrats and “mainstream Republicans” alike, and vice versa for the “far” right.  The “far” left and the “far” right do not have the same scores because they voted against the mainstream (of both parties) at different times.  In other words, the extreme left and the extreme right simply mean that they are non-mainstream factions who have been excluded from the power of agenda even within their own parties and who found themselves opposing the conventional wisdom that the mainstream of both parties agreed on.

That the far left and the far right did not vote together even when they were marginalized could mean that they have little that they agree on and thus have little room to cooperate, but that would be true only if the agenda power of both parties’ “mainstream” wings were not being used negatively, to systematically exclude entire policy areas from being considered at all.  For all we know, there could have been many areas where the far left and the far right could have found themselves on the same side, and between the two of them, muster sufficient votes to seriously threaten the bills, were they pushed forward.  The party leaders, naturally, would not risk such embarrassment, of having their bills shot down by a coalition of strange bedfellows, too often (although “ends against middle” coalitions are not too uncommon in legislative votes generally–just not frequent enough to mix up far left and far right too much in calculating DW-Nominate scores.  NOT that we can tell if the scores ARE in act mixed up, without knowing something beyond the votes themselves.  But we rely on DW-Nominate scores all too often because we know very little about the substance of the legislatures but the recorded voting data is readily available in great quantities nowadays–i.e. the typical “data science” mentality:  we don’t know much, but we got plenty of data that we can use to find patterns.)

At least superficially, there ARE many issues where the far left can work with the Trump administration–on trade, infrastructure, foreign relations, even health care reform.  Many of these issues were either systematically excluded from the Congressional agenda last couple of decades, or presented only in highly truncated form (i.e. Obamacare not including the possibility of expanding Medicare or creating a public option).  We don’t know what kind of possibilities existed from the paths not taken.  If we can indeed abandon the silly restrictive cage of spatial models, that will be worth the while by itself.

Of course, someone like Gabbard makes for an interesting ally to cultivate for a Trump administration.  Trump has been accused of racism and misogyny.  Having a young, accomplished, multicultural woman who is indisputably a patriot in a position of prominence in his administration will make mockery of the accusations from the left while shutting up the obscenely politically incorrect crowd on the right–itself an act of triangulation.  That she should be a well-known leader of the left wing of the Democratic Party will complete the process.  If Trump pulls this off, he will have proven himself a far more capable politician than anyone gave him credit for.

PS.  The point is not so much that there will be cooperation between Trump and the left wing of the Democratic Party, but rather, that it will be with the left wing of the Democratic party rather than the alleged “moderates” whom Trump administration will find easier to work with IF there is any cooperation at all.  It is unlikely that, if Trump does mean to work with Congress and actually get things done, just the Republicans will suffice, if only for the reason that the kind of programs Trump has been proposing are an anathema to the mainstream Republicans.  One possibility, one that is confidently predicted by many pundits, is that Trump will simply lie, pretend that he is getting things done, but in fact do nothing substantive, by surrendering to the mainstream Republicans.  This is not implausible, indeed quite possible or even probable.  But the point I am raising is that he might actually try to do things, and if so, some form of triangulation–albeit of nonspatial variety will be the core.  Of course, that he might want to triangulate does not mean it will happen–it is, quite literally, a three sided game and there is no telling that all three sides can see eye to eye.