A logical fallacy is an invalid element in an argument. There are many kinds of fallacies. For some reason, logicians like to give them Latin names.

One example is cum hoc ergo propter hoc (‘with this, therefore because of this’). This fallacy involves arguing for a causal relationship between two things, simply because they are correlated.

Logicians and politicians have different attitudes towards fallacies. Logicians try hard to avoid them and love to point them out. Politicians often use them, probably because they can be very persuasive. If a fallacy didn’t have intuitive appeal, it wouldn’t be committed often enough to acquire a Latin name.

Journalists also often commit fallacies, usually out of ignorance. Confusing correlation with causation seems to be a favourite of theirs.

A fallacy that may have relevance this week is argumentum ad novitatem (‘appeal to novelty’). This fallacy is committed when a claim is made that a new thing is better than an old one, simply because it’s new.

Like other fallacies, the appeal to novelty has intuitive appeal. People like shiny new things and are biased towards thinking they’re better than old ones.

Two political polls were released last Monday evening. They were the first out since Chris Hipkins’ elevation to the Premiership. In both, the Labour government enjoyed increases in support of about five percentage points.

On Kiwiblog, pollster David Farrar listed the change in support for both major parties in the first poll following each leadership change since 1974. Following 17 of the 20 changes, the relevant party’s support rose. Yet only three of those new leaders went on to win the following election.

Hipkins is smart and affable. Those characteristics have, no doubt, helped win back support for Labour. Hipkins has also promised to wind back some of the government’s less popular policy initiatives.

Whether or not appeal-to-novelty has anything to do with this week’s poll results, Farrar’s data suggest that it often influences voters’ views of new leaders.

Democratic elections work most effectively if people cast their votes rationally. But the pattern of new leaders enjoying an initial rise in support only to go on to lose, is just one of many phenomena that challenge that assumption.

Even so, free elections entail the freedom to vote irrationally. And despite our all-too-human flaws, democracy has yielded the most prosperous societies in history.