Political cleavages are not lines, but bundles

The work of Seymour Lipset and Stein Rokkan described European political systems as the product of several superimposed historical cleavages: between centre and periphery, between towns and countryside, between religious and civil authorities, between holders of capital and wage earners. Each was born of a dated conflict, and each has left lasting traces in the existing parties.

These lines do not overlap. The same person may stand on one side of one and on the other side of the next. It is this non-coincidence that makes political life hard to summarise, and that explains why alliances seem unnatural when one looks at a single cleavage.

New cleavages are added without erasing the old ones: the relation to the environment, to supranational integration, to the globalisation of trade. They are superimposed on the previous ones rather than replacing them, which increases the number of possible combinations.

When several cleavages line up, the conflict hardens: the same people find themselves opponents on every subject, and each disagreement reinforces the others. Political identities become all-encompassing, and compromise on one subject is experienced as a general betrayal.

When cleavages cross, oppositions recompose according to the subject: an opponent on one question becomes an ally on the next. Coalitions are more unstable, but tensions less total, and compromise less costly to consent to.

This is a useful reading key for interpreting a political period without judging it. The question is not who is right, but whether the dividing lines tend to superimpose or to cross, because the two configurations produce public debates of a different nature.

A cleavage becomes structuring only if actors make it salient: parties, media, movements. The disagreement may have existed in the population for a long time without organising political life, for want of an offer to carry it.

This explains a frequent mismatch: subjects that deeply divide citizens do not appear in electoral competition, while others, less divisive in opinion, occupy a central place there. The political agenda is not the mirror of real disagreements.

Looking for one's camp amounts to supposing that the lines coincide. Looking for one's positions, dimension by dimension, makes it possible to note that they do not always coincide, including in oneself. The feeling of being badly represented often comes from there, more than from a lack of political offer.

This implies no judgement about the value of labels. A label is an effective instrument of coordination; it becomes misleading only when it is taken for a complete description.

Faced with a controversy, try to identify how many cleavages it mobilises. A question that mobilises only one is generally settled by negotiation. A question that mobilises three or four at once becomes identity-laden, and technical arguments then have only a marginal effect.

This spotting does not indicate the right answer. It indicates what kind of discussion is possible, which saves one from expecting of a debate what it cannot produce.