I just finished reading the book Early Riders by Robert Drews. I have also been looking up related books and academic articles online and the general conclusion is that there was no military use of horse riding before around 800 BC ...let's say around 1000 BC to be very liberal. Why should this be controversial? For the simple reason that, while military historians and archaeologists in general agree on this fact, Indo-Europeanists who follow Gimbutas do not. She dogmatically pushed for the notion that the IE peoples (I will refer to them as Aryans for shorthand, and I will call the historical Aryans Indo-Aryan or Indic in this thread) spread quite early and that horse-riding was a big reason...she imagined vast Volkerwanderung of tribes, flocks, with men riding horses and conquering Europe and other areas starting as early as 4000 BC and originating in the Pontic-Caspian steppe.
There are a number of reasons this theory had popularity after WW2. First, it rejected the notion of an Aryan homeland in northern Europe (a proposition that indeed has very little to back it up). Second, although it suggested that warfare and conquest was the reason for the expansion, Gimbutas wedded this to two concepts that attenuated or otherwise put in a bad light the notion of a Herrenvolk. On one hand the notion of a diffuse Volkerwanderung or gradual settlement by large populations seemed to "take the edge off" some of the earlier theories of elite dominance, morally speaking; on the other hand, she explicitly wedded this to a feminist fantasy of an essentially pacifist matriarchal Old Europe that was despoiled by horse riding sky-god-worshiping intruders. In a way, having your cake and eating it too.
In the Gimbutas model it is widely assumed that the Kurgan culture is in some ways the original IE culture and that these people spread by their skill with cavalry, which apparently they mastered as early as 4000 BC. But there is no evidence from the steppe for riding this early--no images, statuettes, drawings, bits, etc., nothing; now, if we look to the south of the Caucasus, where there is a stronger material record, we don't find evidence of COMPETENT riding, which would allow for the use of a weapon while riding, until much later. Until that is, after 900 BC or so (there are very few scattered earlier images of riders, but not before 2000 BC, and these show "riders" in very awkward positions...I will talk about this later). Gimbutas' theory and in general the Kurgan horseman theory depends on the idea that somehow people north of the Caucasus mountains were expert riders but that this advance was somehow kept secret from peoples in the Near East, despite the fact that the two had frequent contact (especially in Anatolia). This goes against everything we know about the spread of military technology in the ancient world. The war chariot, once developed (either in eastern Anatolia or in southern Siberia) spread very fast, and the bronze bit, when developed around 1000-900 BC or so, spread within 200 years everywhere from the Near East to Western Europe to Mongolia...with models in Central Asia today still essentially unchanged since the Bronze Age:
ancient Luristan bronze bit:
The previously-used organic bits were inadequate to manipulating the horse...the earlier-still nose rings were the worst, as these could hardly direct the horse to the right or left, and also probably panicked this very nervous and sensitive animal. The point is that as soon as we know it was invented such technology spread very fast, as did the chariot...yet somehow we are to believe that for thousands of years earlier people in the steppe were expert riders. That they were so without the proper tools (metal bits) for manipulating this animal, and so on. And that the inhabitants of the Near East, China, etc., did not learn how to do it despite being in touch with the steppe.
Also consider that the Egyptian Pharaohs used to recruit mercenaries from as far north as Northern Greece. It's not clear why they wouldn't hire the expert riders who are assumed to have then been living on the Danube, nor why these would have preferred to rule poor villagers rather than simply invade the rich Middle East or at least act as mercenaries (this is another of Drews' good points).
When we DO know that good riding developed, around 800 BC we immediately see disruptive predatory raids by various tribes of obscure origins, and then very soon we see a gigantic empire forged by the Medes and Persians, who built it on the strength of their riders. The failure of their empire against the heavy infantry tactics developed by the Greeks and Carians (and that quickly spread to the Mediterranean in general) inaugurated another era of military history, but that takes us too far into the future. The point is that we have no evidence for such expansions by horse-riders before this time, unless we are to take Gimbutas' mere assumptions as evidence. There are other, and much better, explanations for why the Aryans managed to make such spectacular conquests.
Chariots were far more expensive to maintain than cavalry, and also less effective. Yet from 1700-1200 BC we see the unchallenged use of chariots. I would argue, again following Drews, that this is also the time of the earliest Aryan expansions, again through the use of chariot. This is for a future discussion though.