Would you abandon if this were the case?
Maybe I would. One has to trade for or conquer most of the luxuries anyways. It's best to trade for every single luxury that you can during every war.
Is Republic always the right answer? Can Monarchy be the right answer where science is not an issue?
Monarchy could mean more content citizens early due to less war weariness. However, once the conquering phase is over, as many happy citizens as possible and specialists will do the most for score. The luxury slider will have more of an effect in a Republic after the conquering phase. If your empire remains a Monarchy, then you don't have as much scoring potential. If you revolt a 2nd time, then you have a food loss from the standard tile penalty ("the despotism penalty").
As Bartleby suggests, 60% (maximum land, minimum water) archipelago takes longer. I tried to load Darkness's final save using CrPViewer and it tells me something about a multiplayer setting. What is the best conquest on some 80% pangea map is not the same for some 60% archipelago map.
Do you want to start building an army before you stop building settlers for "natural" expansion - i.e., conquest over natural expansion?
I haven't personally when playing without roaming barbarians or higher. I've more looked at trying to do something to increase score. For core cities, I've highly preferred building marketplaces over training soldiers. I think only when I did Conquest games with the Iroquois on Deity did I actually train an army of fighting units to conquer aggressively over building marketplaces. Marketplaces increase score as soon as built and have a lasting effect towards happiness (unlike temples, cathedrals, and colosseums which only give content citizens). They also delay any rioting from happening and thus help with the war weariness issue for Huge maps.
but do you play Histo at higher levels as a conquest game first and leave the expansion for later?
No, no, no, no. Whoever told you it more resembles a conquest game first I don't trust on this point, unless we're just talking about preventing a loss of the game. Given that you want the most score you can get from that map, it's more like a domination game first when thinking about war. Score increases from conquering of cities come from more tiles in one's empire cultural borders or from new happy/content citizens. Or from having settlers replace cities. But the later can easily be slower for population growth at high levels than quelling resistance of captured cities. Fortunately, native nationalities of some citizens can still exist in 2050 AD.
My advice goes that when you have decent enough happy and content citizens to handle war in a Republic, then you push for going up near to the domination limit (above 60%, close to 66%, but never over 66%) through conquering. Captured cities can put out workers or settlers in a reasonable amount of time, and help with getting more territory. I have the settings set to "always build previous unit", and that helps once I change captured cities to workers/settlers.
You also want to get to certain techs as quickly as possible especially during those periods where you are doing self-research and this would require Republic.
You have a good point here. Map Making and Navigation (for luxury trading if you don't have The Great Lighthouse or have to traverse an ocean for enough luxuries), and Steam Power (for more production in the core from rails, faster troop and worker movement, and growth in captured cities). Maybe I didn't need to say that, and of course, I don't think I need to mention others.