Political Parties Are Bad at Self-Examination. We Shouldn’t Imitate Them.

Last week, as Abdul El-Sayed clinched Michigan's Democratic Senate primary by well under a percentage point, political commentator Matthew Yglesias made a prediction on X. El-Sayed, he wrote, would likely win the general election in November—but by a smaller margin than Jocelyn Benson is expected to win her race in the same state, smaller than Democrat Rebecca Crowley's margin in Wisconsin, and certainly smaller than Josh Shapiro's in Pennsylvania. Then came the real point, delivered as a dry aside: everyone on the left, he predicted, would draw "intellectually honest, rigorous conclusions" from this.

The sarcasm was obvious. Yglesias wasn't really forecasting an honest reckoning. He was forecasting its absence—betting that nobody on the left would learn anything useful from the election results.

Yglesias's observation does not apply uniquely to the Democratic Party. It expresses an almost universal truth: institutions that lose, underperform, or narrowly survive a warning shot tend to avoid an honest reckoning.

The clearest case study is the Republican National Committee's 2013 "Growth and Opportunity Project," universally known as the RNC autopsy. Commissioned by chairman Reince Priebus after Mitt Romney's 2012 loss, it was, by the standards of political self-examination, genuinely unsparing. The report stated plainly that the party's message was weak, its outreach insufficient, and its primary process broken. Most pointedly, it warned that the GOP needed to stop talking only to itself and embrace comprehensive immigration reform or risk permanent marginalization.

Commentators at the time called the report stunning and hard-headed. It was also almost entirely ignored by the GOP. The diagnosis might have been right, but the patient simply refused the treatment.

The Democratic Party's story is not so different. After 2016, arguments about what had gone wrong—economic anxiety versus identity politics, message versus turnout—were fought loudly but never resolved into anything resembling consensus or correction. The same unresolved argument resurfaced after 2024, and by every visible sign, it is resurfacing again now, in real time, around a Michigan primary result that landed within a rounding error.

Yglesias's joke works because everyone recognizes the pattern: a party treats a near-loss as a mandate rather than a warning because acknowledging the warning would require admitting something uncomfortable about itself.

What both cases share is not stupidity. The RNC's report was smart. Democratic strategists are not naive about the weaknesses in their coalition. What both share is a particular kind of evasion—one that Jewish tradition identifies as the opposite of a practice it considers essential: cheshbon nefesh, literally "an accounting of the soul." Without regular self-examination, Jewish sages teach, human beings are prone to denial and self-deception, with potentially devastating consequences.

Rabbeinu Yonah, in the opening of Sha'arei Teshuvah, describes self-examination not as an occasional gesture but as a basic requirement of moral life—something a person must return to continually, because the alternative is drift. The Ramchal, in the first chapter of Mesillat Yesharim, makes an even sharper point: people fail not because the truth is hidden from them, but because they do not stop long enough to think about it. Habit and self-interest crowd out reflection before it can even begin.

That is precisely the mechanism at work in too many institutions, including political parties. Yglesias's joke is a simple description of reality: when you avoid cheshbon nefesh, you draw only the conclusions most convenient to you.

This is also why the month of Elul has the particular character it does. The Jewish calendar does not ask for reckoning only after catastrophe. It builds in a month, every year, whose purpose is to force the pause before drift becomes irreversible.

The Rambam, in Hilchot Teshuvah (2:1–2), sets the bar for us: real teshuvah is not regret or ritual language. It is demonstrated when a person faces the same circumstance again and chooses differently. A party that wins very narrowly, or commissions a report, and then returns to the same old behaviors has failed this test. It has performed the political equivalent of vidui (confession) without changing its behavior—confession as performance rather than as a turning point.

Political parties do not observe Elul. But they do have something functionally similar: election results that arrive, with some regularity, as a knock on the door. That they fail to take advantage of that opportunity is regrettable and even alarming, but at least it teaches us the price of such avoidance.

The Jewish tradition's insistence on cheshbon nefesh is, among other things, an attempt to prevent that avoidance—a reminder that truth can't be reached without honest self-examination and a willingness to sit with discomfort. Political parties might not know how to do that, but Jews have been practicing it successfully for thousands of years.