From my perspective, no it is not a hard science. I believe that the scientific method has quite specific parameters, and we should always acknolwedge and work with those parameters. They include methodological naturalism, a commitment to cause-effect analysis, repeat-testability, and so on. Historical study, in particular, is not subject to repeat-testability unless you want to run large, society-wide, multi-generational unethical testing.

The great problems I encounter are usually not within the discipline itself, but communicating and engaging with non-historians. One of the great tragedies is the rise of philosophical "scientism" which I would describe as an ideological commitment to science as both the only true epistemological method (if it's not science, it can't be known), and the last great hope for humanity (science will solve our problems).

Just because history isn't a hard science doesn't mean it doesn't have rules, logic, procedures, rationality, etc.. There is good history and there is /r/badhistory. History doesn't need to be science to be valid inquiry.

I do recognise that some sub-disciplines are scientific, or employ science, for example some form of archaeology. However, in terms of studying documentary evidence and formulating hypotheses and writing accounts, science is not the right categorisation.

Lastly, to tangentially comment on the link you posted which I confess I only skimmed because it became boring quickly. History isn't really in the prediction business. The old "condemned to repeat the past" is, in my view, a very old Hellenistic view of history as cyclical that I am not sure is true. History may have some predictive powers, but Historians' work isn't as fortunetellers.

Whether history is a science can certainly be debated, and it often is. The world "history" comes from Greece, where it literally meant "inquiry" or "investigation." It has evolved today of course, but it's accepted by most that history is the "search for truth."^1

Herodotus is credited with being the father of history, and Collingwood claims he was responsible for the “conversion of legend writing into the science of history.”^2 History shares a lot with sciences like chemistry and physics. The scientific method for example, you make a hypothesis, support it with evidence, and then come to conclusions. This is the same as in any historical journal as it is in a scientific one. Theories without evidence are seen as bad history, the same as theories without evidence are seen as bad science.

The biggest argument against this theory is that sciences deal with natural laws, but the positivist school of historians believe in "regularities in human behaviour."^3 They believe that historians work with general laws too, in assumptions about human behaviour. This theory is that historians use counterfactuals, the concept of ‘if’. The often quote example comes from Granet. Granet claims that ‘if’ the ancient Chinese did not know the hibernation rituals of animals, then they would have had less rigid rules about their own winter conduct.^4 While the argument isn't perfect, it can certainly be accepted that historians work with general laws - that events have causes and consequences, that people act on motivations, and that invading Russia in winter is folly!

There are a plethora of arguments against this theory of course, it's something I could talk about all day, it's a truly fascinating subject. I'll just put forward my view however, history does share similarities with science, like it shares similarities with humanities, but I personally believe history is quite unique, history works with the scientific method, but what distinguishes it is the presentation. History is presented through prose, through literature. History can be truly beautiful. The art of effective arguments and convincing rhetoric. Entwining research with new information, and putting it forward in a manner that can be entertaining to the reader. I cannot recommend Trevelyan’s book Clio, A Muse enough, it's on this subject and is a truly brilliant work.

^1 A. Marwick, The Nature of History (1970)

^2 R. Collingwood, The Idea of History (1956)

^3 D. Bebbington, Patterns in History (1979)

^4 W. Todd, History as Applied Science (1972)

A key element of "hard" science is experimentation to confirm or discredit a priori hypotheses. This is impossible for history (the discipline), because the past is irretrievably gone. We historians cannot build a "historical particle accelerator" to test whether it was Hitler's mismanagement of the Eastern front, or Allied economic superiority that was the decisive variable in the eventual outcome of the Second World War.

The philosopher R.G. Collingwood compares the work and methodology of the historian to the detective (see his The Idea of History), whose task it is to reconstruct the past based on the limited evidence available. History is ultimately inference to the best explanation.

As a socialist, I wouldn't call history a hard science, not in sense that chemistry or physics are. That said, history can (and should) be viewed as "scientific" in the sense that a historical hypothesis should be created using a "scientific" method, i.e. conclusions should be drawn from observing the physical world; in the case of history, the evidence available to us in the modern world.