Um bom livro sobre filosofia da história, isto é, problemas filosóficos que aparecem quando estamos lidando com a disciplina da história, de um bom capítulo mostrando a insuficiência tanto das correntes marxistas mostrarem a superioridade do social sobre o individual e daquelas individualistas (como a de popper) de mostrar o contrário. Ademais, trabalha com paciência e perseverança sobre a possibilidade de termos conhecimento sobre o passado, a necessidade de levarmos a sério a temporalidade do futuro (e, portanto, sua abertura, contra o determinismo), o trabalho do historiador como estabelecendo narrativas (explicando a mudança entre dois eventos passados, a partir dos fatos estabelecidos no presente), a significação histórica portanto surgindo dessa mirada, de modo que o passado mesmo não possa ser conhecido em sua totalidade (dado que sua significação fica sempre reaberta a reavaliações futuras, narrativas que se tornarão significativas ou insignificantes). ( )

Philosophy of history hasn't exactly been a very lively field of research in the late 20th and early 21st centuries, so this attempt to apply rigorous analytical analysis to the subject feels like a new start even though it was written in 1964. But analytical rigour has its pros and cons, and the book takes a bit of a nosedive in the first few chapters when the authors sets out to investigate completely inane questions such as (1) whether all statements about the past are meaningless, and (2) whether or not it makes sense to assume that historical statements have no reference, because the world might, for example, have been created 5 minutes ago. It's a competent philosophical analysis as far as it goes, but I don't think these bizarre questions really merit this much thought.

So any historians reading this book would surely have quit during the first six chapters out of pure frustration, but fortunately the author switches to a much more interesting track in chapters seven to nine. Here he introduces the Universal Chronicler which carefully takes note of everything that happens in the world exactly as it is happening. The author's basic argument is that the Universal Chronicler is not an ideal historian. If a given event occurred at time t, historical narration and explanation of that event has to make reference to subsequent events occurring after the time t if it is to make sense. Historians should therefore not lament the difficulty of studying temporally distant events. Temporal distance is necessary for historical understanding. It's an interesting analysis particularly because it adds the historian herself to the chain of effects that resulted from the event. If the historian's work influences the present, a complete historical narrative of the event would in the future have to include that work.

In the final three chapters the author again veers off into poorer pastures. He undertakes the task of formulating a general account of historical explanation. He discusses at particularly great length Hempel's work on causal explanation in physics and tries to find common ground in historical explanation. This attempt feels quite outdated. The paucity of insights in this analysis illustrates why modern philosophy of social science has moved away from unfruitful parallels to natural science. So in conclusion, the middle part of the book contains an argument worth contemplating. The beginning and the end can be skipped in good conscience.