Talking to Jean-Maxime Moris and Jeff Spock, Amplitude's executive producer and narrative director on Humankind, I can't help but feel a little bit guilty for immediately bringing up Civilization. There are more 4X games than Civ - and older ones at that - which means comparing every new 4X game to it can feel more than a little trite. But Sid Meier's influence is the one that lingers. It's also the closest, by far, to what Endless Space and Endless Legend developer Amplitude is aiming for with Humankind: a historical, diverse, and deeply optimistic game about the miraculous progress of the human race.
It's not perfect - certainly not yet, as the build I played was still waiting for a number of pretty crucial systems to be finalised, or even implemented at all - but what makes Humankind immediately stand out from its illustrious cousin is its approach to one of the genre's biggest frustrations. Humankind is trying to solve the problem of culture, that weird and conceptually squishy amalgamation which feels essential to any game about human history but has, so far, proved to be a bane of the Civilization series and others like it - and the team at Amplitude's Parisian studio might actually be onto something.
Amplitude's approach with Humankind is to make you choose a new culture for your civilization each time you progress into a new era. You don't choose a civilization at the start and play with them throughout, and instead all start from the same blank slate - with a customisable avatar representing you as a sort of detached 'leader' through the ages. The scenario I played lasted a couple of hours, up to a maximum of 60 turns or two eras, from more or less the very beginning of the game, which meant I was able to get a decent sense of how the culture-hopping worked out. The immediate thing you notice is how it maps much more sensibly to actual human history. Societies - speaking very generally here - tend to evolve according to the natural and social surroundings, so ones with fewer military rivals nearby and lots of lush, arable land might move more towards a peaceful, agricultural society, such as the Harappans (or Indus Valley Civilisation), which were based in what is now north-east Afghanistan, Pakistan, and north-western India in around 3300 - 1300 BCE.
On the other hand, you might be surrounded by expansionist rivals and have access to early metals like bronze, such as the Mycenaeans. In Humankind that is essentially the logic you follow as a player: after some early exploration you'll settle, and then have a choice between the Harappans, Mycenaeans, Egyptians, and Babylonians, which lean towards the specialisms of food, military, production and science respectively - and if you want to get anywhere you'll need to think about how that choice ties into your own situation at hand. Each of these also comes with a legacy trait, which is an ability that this culture allows you to keep using throughout the game, an emblematic quarter, which is a unique city extension (basically a district, if you played Civ 6), and an emblematic unit unique to them too. Later on, as you advance to the next era, the choices widen - similarly to how the number of different established civilizations widen in history as you progress through the years - and as such so do the specialisations within them, with things like commerce, order, and aesthetics coming into play.
Part of this, executive producer Jean-Maxime Moris explained to me, is simply a case of giving you more things to play with. "It is a much more dynamic approach to history, I would say. You could argue that it doesn't make sense to have the Olmecs with the Huns [in the same region], but it's part of the fantasy... what if you had access to the most renowned cultures of the world and you were able to build your 'super culture', your civilization with them?"
Narrative director Jeff Spock echoed that, when I teased about a certain rival-that-must-not-be-named. "When I play, for instance, a different historical strategy 4X game, basically you choose a culture to begin with and you choose a civilization, and that means you've chosen your gameplay and your victory condition, and you're on rails. And if for some reason somebody else gets something halfway through [before you], it's like 'Oh, well, I just wasted 20 hours playing the game, I have to start over again'. What I love about what we're doing here is you can get halfway through a playthrough where you wanted to be the Science Guy, but [then say] 'You know what, it looks like I'm maybe gonna go industrial, or aesthetic', because I'm not going to make it there with Science. You can change horses mid stream and still make the other bank."
The other side of it is more philosophical - more about that point of making a game of human history more like human history itself. "From my point of view," Spock said, "I think there's two reasons I feel pretty strongly about it. One is: here I am, an American sitting in Paris, married to a French wife with two French kids, and my ancestry is a little of everything from North America and Europe. And yet every country that's here today is this melting pot of what came beforehand - there are cases like Japan, which is an island nation that shut itself off for a couple of centuries, that sort of thing - but basically, humanity is this big, big ugly mix of everything that came before. And I think a game that says, 'you're this people who play this people for 5000 years' - it's not realistic. It doesn't reflect history. It doesn't reflect diversity. And so I was happy to step away from that.
"We can actually now portray, in the game, cultures like the Harappans and the Olmec, who we know existed, we know they did amazing things, we can see the runes - but there's no names. We can't name the cities, we can't name the people, we can't name the rulers, but we know they existed and we know they were amazing. If you're going to do a game where you play them for the whole game, and they have a leader, we don't have enough information, literally, to do that. Whereas if it's part of a building block, you can take these otherwise ignored cultures and integrate them completely logically into the game flow and say, 'Jean-Max, show us what the Harappans could have done in the ancient era'. I find that kind of exciting, and kind of philosophically very pleasing also."
What makes this kind of shift in approaches possible is Humankind's other big gamble. There's only one victory condition in the game, which is having the most Fame, itself earned by collecting Era Stars, effectively a type of reward given for completing specific objectives over the course of the match. These also gate your progress from one era to the next. So for example in my early-game playthrough, I started with a single Nomadic Tribe in the Neolithic Era. To advance to the Ancient Era (where I got to choose my first culture), I needed to get a certain number of Era Stars, with 'accrue 25 Knowledge' and 'gather 4 units' being the two objectives given for earning them. You can also pick up Era Stars for specific feats as you go - being the first to build an artificial wonder, say - and for cumulative tasks, like earning a star for winning a certain number of military victories during an era, then another for a greater number of victories, and so on. In other words, you're naturally rewarded for excelling within your specialism, like that example of military victories, but specialising doesn't hinder your ability to pick up stars elsewhere.
Again, this comes down to a specific approach to history from Amplitude, as well as the benefits it grants to gameplay. "We wanted to improve the way we're looking at history," Moris told me. "It's about what you are remembered for, whether positive or negative. So you might chart an entire continent, discover writing and invent the wheel in the ancient era. And you might get crushed in era three in a battle, but what you did in era one struck the minds of the people so much, it actually gets you enough fame to win the game. So just because you lose a battle doesn't mean you lose the game.
"It may sound or look like it's very simplified, the way we boiled it down to one score, and that is true, but what we managed to get rid of in terms of complexity, we made sure we didn't lose in terms of depth... the world deeds and the era stars are crystal clear on that interface, very easy to read. But then how to go about achieving them, in which order, depending on what everyone is doing is still very, very, very deep and satisfying."
It's a little early to say how satisfying that truly is - Humankind's later eras themselves are still too far under construction for us to have had a look at, for a start - but the turn-to-turn stuff of the first few eras was certainly deep. In terms of units, Humankind has a kind of stackable system, where a single 'army' on the map can be made of up to four units within it - our Nomadic Tribe in that first Neolithic Era was actually two tribes combined. Knowing when to clump and split them will be important, not just militarily but in early game min-maxing as you juggle exploration with survivability. This is also governed by a generals system: each unit needs a general to go it alone, so there'll be times when you're restricted to stacking them in one army because you're out of generals for commanding them.
Rather than looking to found a city immediately, the first few turns are, fittingly, quite nomadic. You'll roam about looking for Curiosities - 'goodie hut' equivalents, to Civ fans - that grant you a chunk of Knowledge or Food, allowing you to expand your unit numbers and advance to the next era. Founding a city happens once you advance to that next era and choose a culture, and then it's more land-grabbing mixed with some more considered specialisation.
In terms of acquiring land, things are surprisingly money-focused, which felt a tad anachronistic, especially when held up against Amplitude's emphasis on mirroring history in other areas. The map, which is a gorgeous, impressionist thing, is pre-divided into regions, with lots of tiles within each. Planting an outpost tower on a region will make it yours - meaning no other players can build on it without destroying your tower - and if you have a city in an adjacent region that tower will bring it, and its resources, under that city's control. Those towers can then be upgraded to cities, for a hefty chunk of gold, which is where things feel a little strange given most societies haven't unlocked things like currency at that point, even from the tech tree in-game. You can also buy out productions with gold too, to rush them through.
Cities themselves are made up of a central structure, with extensions acting like districts, taking up tiles around it, and other city upgrades like infrastructures not requiring space at all. Each city has a population, which can be auto-assigned to different specialisms or manually assigned things to generate. Again, if you're familiar with Civ, you'll notice that the cities are where things are by far the most similar, with differences only really coming in terms of slightly different terminologies and tweaks.
Where things branch off again is combat, which is the final notable twist on the 4X formula. Instead of the surface-level battles that effectively auto-resolve in games like Civilization, or the fully tactical, XCOM-like turn-based ones on separate maps that you get in an Age of Wonders, Humankind has a kind of turn-based, on-map hybrid. "What happened was we got an awful lot of requests - I wouldn't go as far to say pressure - but let's say 'strong interest' from the community, to make it turn-based, because in Endless Legend we had a certain level of tactical choice in the battles," Spock told me. "But then it was sort of running on autopilot, and people kept coming back and saying we want full tactical, full tactical, so it was a big step to take, and it's maybe a risky one, but we felt there was so much demand for it out there that we just said OK, we're going with that."
Like most feature suggestions from fans, it's an interesting idea that really only half-works in practice. When units fight each other, an area around them is blocked out on the map and you enter a turn-based battle. The aggressor has to either wipe out all enemy units in three turns, or capture a little flagged point that's marked on one of the tiles, which the defender has to try and hold. The issue I had is that, for one, three turns feels very slim, in terms of allowing you enough time to do either, and also that these battles seem overly lethal. Every conflict leaves one side's army either entirely wiped out or very nearly depleted - which is significant when it could be a stack of four you just lost, and how much time or money it can cost to produce each one. In the demo I played, there was also - thankfully confirmed after the fact as at least partly a bug - a problem with retreating, where running from an enemy attack would see your unit cheese it across at least ten tiles of the map in a random direction, and the enemy would always catch them the next turn anyway. It wasn't clear why you'd retreat in the first place when you can always be caught, and the randomness itself means map positioning feels a little pointless. In later stages, neutral, effectively barbarian chariots were wreaking havoc on my empire, appearing from off-screen in the fog of war and travelling huge distances through my land to attack me in different places - it's not clear how you can really stop that, in terms of positioning or defending a specific spot.
There's also just a huge amount of the game that's still a work in progress. Religion, civics, ideologies, diplomacy, and much of the UI - especially in combat - are all features still in the works. Some of these sound very promising, such as civics and ideologies, which Spock described to me as "very organic". You could end up, for instance, "having conflicts where you, as the avatar leader of your civilization, might want to go to war against somebody or might want to have peace with somebody, but because of the civics you've chosen, or the religion that your people follow, they won't be happy with that." There'll be what he called "downstream effects", basically knock-on effects of past decisions and specialisations that mean you have to approach things in later eras in a different way, which could dovetail quite nicely with the more open approach to switching between cultures and specialisms themselves.
As always then, it's a wait-and-see thing, with effectively half of the game's systems still up in the air. But wobbly combat aside, most of what I played was genuinely exciting. 4X games live and die, for me, by their own philosophies. What makes them interesting, and what makes them actually fun, is the way they simulate not just the feeling of being some all-powerful emperor, but an actual state of mind. They simulate perspective, basically, on humanity as a whole. Humankind's seems to be one of multiculturalism, nuance, and above all optimism, and personally that's something I couldn't be more ready for.