You're clearly thinking hard enough about this (for which thank you, BTW) to realize this is an exceedingly broad question.

The answer you quote at length from the thread on Diamond reflects a common position in anthropology and related social sciences (sociology, geography, history) but it is not a universal one. I respect where that person is coming from but I do not entirely agree with their opposition to generalizing theory. This is in fact a basic tension in the social sciences, between particularism and the urge to identify broader causal mechanisms or patterns.

The founding figure of American anthropology, Frans Boss (himself German and a physicist by training), put the tension in these terms:

"We find in anthropology two distinct methods...the historical method which endeavors to reconstruct the actual history of mankind, and the generalizing method which attempts to establish the laws of its development" (Boas 1904)

Instead of getting too into the weeds here (I'm on my phone), let me point you in a couple more general directions.

Start by reading Bruce Trigger's magnificent History of Archaeological Thought. This will walk you through all of the arguments about what we can know and how we can know it about the past, about the "little picture" issues like how to interpret a scatter of artifacts in behavioral terms and "big picture" issues like how societies change through time and why.

Once you've done that, dig into the archaeological literature on "complex society" or "social/cultural complexity," which is the realm of science most cconcerned with how societies become more complex through time, why, and why they then fall apart. Include "archaeology" as they keyword because it'll weed out a lot of the Harari types who don't know what they don't know

Key thinkers here (roughly in the order I like their work, others will Have Opinions) include:

Joseph Tainter Kent Flannery (especially the 70s stuff) David Wengrow Ian Morris (his articles and chapters are much better than his books, IMO)

Hope that sets you off in the right direction, good hunting!

I would actually challenge the notion that historians don't or can't make strong, explicit causal claims. In actuality, there are plenty of ironclad causal claims that can be made about history, they just tend not to be causal claims that are sufficiently interesting or important enough, e.g. trivial, or otherwise even useful to apply to a matter of interest. Or at least alone they don't answer the questions of interest. Let me explain.

What you're actually talking about are causal claims of sufficient complexity and universal scope to explain a phenomenon of interest. In the field of naturalistic science, from the perspectives of falsifiability there are established standards for scientific laws that make it desirable for scientific laws to trend towards the most universal and efficacious construction of that law, as opposed to less universal forms. Why? Because they are more falsifiable. A causal claim for a force of physics that applies only to a specific planet is less universal and less falsifiable than one that applies universally to all astral bodies. This specificity is undesirable from the standpoint of scientific activity, since the universal one is more amenable to being falsified (given its vast applicability and the number/ease of counter examples that could falsify it). Being more falsifiable (in this case via univeral applicability) is good if the law withstands many attempts to falsify it (in other words, it is a stronger law even than another similarly unfalsified law that happens to be more circumscribed).

In the case of historical evidence we can assemble a reasonable degree of evidence to assure the causal link between some set of causes and certain effects. Some of these indeed can be ironclad, but for them to be ironclad they almost have to retreat to being so specific that they lose general applicability (such as historical person 1 being motivated to do a specific thing in response to cause A based on their own admission of this being so). A journal entry might be sufficient proof of that. However as soon as you ascend to say, all persons of shared category 1, then already you almost have an insurmountable burden of evidence to directly prove they did some thing for some reason (such as all blacksmiths switched to iron from bronze because iron was better). So the claim gets walked back in relation to scope, complexity, and availability of evidnece for the matter being studied.

Like naturalistic sciences, history is empirical in that it relies on observation, evidence, data. It is also rational in that it relies on reasoning to construct perspectives and narratives on past events. However there are several factors that necessarily distinguish the study of history from that of naturalistic science. For one thing, direct empirical observation is not possible (outside the scope of persons life perspective) nor does direct experimentation have any bearing on historical claims. Instead, limited evidence is assembled and reasoned about, and cases and examples are used to deduce possible claims. While we can certainly rationally arrive at strong links between causes and certain effects, they just tend not to be satisfactory to the listeners interest, because the ironclad claims either wont be sufficiently universal, theyll be trivial and unimportant, or they are so qualified as to be unsatisfactory (to some). So we can reasonably make the causal claim, for instance, that throughout history any person moving to any urban space will encounter more people than they would otherwise in a rural space. But is that of sufficient interest for someone wanting to casually link a more specific effect, such as being more likely to die due to communicable disease to this same phenomenon of urbanization?

The complexity here explodes into so many dimensions that ultimately causal certainty often escape our investigations about matters of interest. For instance, we could reasonable argue that it indeed does increase risk of disease. But we have to add many qualifiers for that to ascend to a large scope for the claim, since it may not be for every urban resident in every urban society, in every era or even in every city in the same region that this causal relationship exists. So all the sudden we've gone from a strong causal statement to a more qualified one, just to achieve greater scope. And this often ends up unsatisfying when it comes to the desired "hard answers". Meanwhile a more trivial causal claim would be all new urban residents must find food in their new urban space. Its admissable as 'true' and there is a causal claim, but unless you situate it to more important factors is not of great interest, not even to argue against much less claim in the first place.

And this is all assuming that evidence is even available, or that it all lines up and rationally follows from a sound argument, not to mention different, equally rational interpretations that reach different conclusions from the same evidence. Hence, why you'll always easily find that historians disagree on such explanatory factors and phenomenon for even seemingly simple matters, let alone the massive almost impossible scope of the questions asked in the popular narratives of history referenced.

All this is to say that there are many strong and defensible causal claims in history. They just tend not to be the ones that people are most interested about. Indeed to make an interesting causal claim almost requires that the historian is willing to step into a murky, difficult and contestable territory, while 'settled' causal associations are not even mentioned since presumably theyre nor worth arguing for. Either way, if you were only making ironclad causal claims you'd likely not be writing about anything worth investigating or paying attention to. The contestability is a feature, not a bug, of historical discourse!