Truly alternative histories would require the designers to flex some muscles in the domain of anthropology where.... the knowledge just doesn't exist, after a point.
It's what I've always deplored in History books : History is basically saying what happened, not understanding things. I'm pretty sure we can find reasons to what happened, systems, phenomena, etc... I'd rather think there is a scientific branch which is basically that, but i don't know its name unfortunately.
Like, imagine if Humans just figured out a republic quicker because they tried really hard (in place of other developments). How the frick could anyone know how to write what happens at
that point? This is what I think the "take away the tech tree" notion becomes.
Actually there were a republic earlier, it's called greek democracy, and more generally greek golden age. Did you know that someone already designed the plans of a steam machine back then ?
I think the thing that prevented him to realize it is that technics were not mature enough.
There is some logical evolution of technics, that I talked about in a topic here : mostly, technics improve, or so is told. For example, you can't create an object that requires a 5000°C temperature with two tree branches alone : you have to elaborate your technic. Like building a furnace that can resist the heat. Most of the time, those improvements would require a global understanding of physics, which is never reach even today, so those improvements are made by guess, try and fail. That's why they take so much time and seem so much random - which I think they are.
However, there can be inventions that truly require some understanding, even though they need intuition also. If somebody would have tried to build a battery in antiquity, without any understanding of physics, he would have been considered totally mad. And yet ! The guy who discovered electricity and artificial light should have look like totally crazy, even for kinfolk, wrapping copper wires like crazy. (we can't do it by accident in prehistorical eras for example, we NEED some basic understanding)
So technics are generally "nesting", either with science or another technic. That's why techs in Civ can't be totally random. However it would be interesting to try to analyze what depends of what, although I fear that science History is mainly a story of chance. We can for example perfectly imagine a war machine operator doing a miss move when charging his machine : there's a wire of metal somewhere, it breaks, it wraps around something, forms a circuits with another part, and *poosh* it shines. After, the operator is proceeding by elimination to see what parts are essential to the system. (removes random parts, and seeing if the light is still there) *poof* Edison.
But this example is imperfect : you need war machines. Can we imagine a situation where electricity is discovered by a Neanderthal ? Probably, although he would need to manipulate something. (invented something else or trying to invent something, which proves he is crazy or making the link between his direct needs and something that doesn't exist yet)
But such link would be hard to do in prehistory. I believe that it was possible because of a thought stream that stipulated that "magic" is physically possible in our world because it obeys to its rules. A non-magic magic. A wonder. Science.
All in all I believe that there's a general direction in technics, but that it is not so much technical. It's just mostly and highly probabilistic, a thing calling for another, and anthropological.
Less determinism... well, you could make barbarians raging as the default, ending whole games because you didn't make that 8th archer in paranoia before trying to improve a tile.
---
Having a tech tree, random research? Is a thing. SMAC sorta was fair with that, it's possible. Another thing could be Great People theory, where literally only a Great Thinker can discover a tech, but that's just reframing the mechanism not creating nondeterminism, because now you're just triggering the GP rather than getting beakers; you gotta work out where that number comes from.
Civ5 choosed to make it depend highly from number of people directly. That's a theory, that I thought out earlier myself. But i don't think it's realistically accurate. Especially, it makes bigger always better, whereas it can be true for "developped" countries, yet not all countries have been "developped" at the same time, and yet stays to be determined what is "developped", which is not counted in rough happiness for example, or then a precise concept of it (ideology), lefting behind other concepts yet the absence of concept (which is probably better, because happiness can't be conceptualized and comes from a various set of factors, perception included).
So a good system for science in Civ could be any so that science can fluctuate significantly despite the size of the civs. I would personnally add an ability for the player to make it fluctuate significantly and quickly, at will (example : Civ1/2/3/4 sliders) or step by step (example : building/rushing libraries).
I still see two core problems to solve with Civ no matter what your ideals are.
One, the rival problem. Your only opposition are your rivals, who play the same as you, and the barbs. You know that every rival is a true threat to your destruction and you can attack and harry them, and also, you know they can do whatever it is you're doing right. I mean this in contrast to , say, the Rome case. Rome was small. It became gargantuan. Is it not certain this was possible because everybody who could have quashed Rome had bigger fish to fry - each other? And is it not also at least, a little bit, owed to some irrationality and maybe racism? Discounting an upstart nation because you think in those terms, because your political constituency thinks in those terms, not in terms of "The Rome player is getting bigger. They -will- eventually become a rival." ?
Two, the conquest fairness problem. You let players attack each other. And you have to allow both fights for land rights and chastisement, and fights for existence. You must allow players to check each other with these tools, and to express these elements (*this isn't a simulationist or gameist issue but rather an artistic one of the worth of the game as an artifact). But you also have your war vs. peace development, industry being more profitable.
So how to evade, at all, that you will then have the situation of a player who is hindered merely because he contends with an irrational buffoon who attacks where he cannot win? You war and war, and now you're both irrelevant. So in formal terms, the issue is: If you did nothing wrong, you shouldn't lose. Neighbouring a warmad lunatic is nothing you did. Something you didn't do isn't something you did wrong. So neighbouring someone who wars irrationally must not imply that you lose. So there is some correct move for when your neighbour wars irrationally.
Your correct move cannot be to destroy him, by hypothesis and symmetry. But you can't compel him to play well. Your option cannot be the peaceful development the other player so irrationally fears, because then it is -literally- not right to war, so just... start the game in a later era by default, why don't you.
By process of elimination, the game design must envelop a third alternative, not military buildup nor industry, but something to counter a stalemate, which in turn must not dominate either other option. (And of course, it should be hard to accomplish, so that stalemates sometimes happen as an indication of skill or lack, which -would- justify players elsewhere winning.... except it has to not be harder than it is to irrationally benefit from CivCity-er neighbours.)
You have to turn a profit from enduring the irrational attack; punishing the other guy's mistake is not enough. That's fine and fair - something easier than successfully working out federated peace, but less profitable than such; cost and reward are balanced there. But measuring "irrational war"? Oogh. My best guess is reverse war-weariness. That's all I've come up with in like 100 hours of thinking about it.
These core difficulties unresolved, a real Rise and Fall situation could never be fair or fun in anyone's palate, even far from the strategy and competitive player, I think.
It's a rather odd way to turn out the problem.
I understand your first objection, Civ is a game with victory conditions and acts towards them that put the focus on enemies. But that's one of my main reproaches to Civ5 : everything seems so much gamey, there is a pre-set number of normal civs in each games, you encounter those civs rather early, trade with them, do wars, etc... it's like the U.N.O. has been set in -4000 BC, especially with the ridiculous and debatable "denounce" feature. So it's possible to temper this. civ5 is also a pretty much "streamlined" game, with some lack of features that could divert temporarily the player from his "military enemies". Ideally the game would marginally be about empire size/conquest. That would need (a lot) finest mechanics. Not sure it is the philosophy of the devs, only by seeing how a Civ5 late map is cluttered (with a lot and extremely simple elements, the complexity being "spread out") says long.
Now I don't see how "fighting against a fool" can be hindering. Vs AIs that never happens because they make peace quite easily. Vs players it can happen, but then you didn't choose well your game mates, or can convince them by chat that's it's better to stop war and maybe also ally. I'd rather think it's an episode from your playing that you didn't know how to deal with.
And how can you know they can't beat you if they don't try you ? 
