This image will make you understand American history
Introducing the historigraph: A map of space and time
This is a visualization of American history:
Look at it. Study it. Download the full-size image here. If it feels overwhelming, relax. I’m going to walk you through it.
This is going to turn that giant cloud of terms, dates, names, and amorphous concepts in your head that you call your knowledge of American history into a crisp mental model that you can actually use. I know this because that’s what it did for me.
Using only the winning party in each state for America’s presidential elections, this graphic, which I call a historigraph, visualizes our nation's convoluted political history. How can it do that with just elections? Simple: it uses all the stuff you already know. That’s the cloud of terms I was talking about.
The average American’s perception of our nation’s history probably looks something like this:
It’s unstructured, impressionistic, and full of gaps. Others may have a bigger, denser cloud of terms without as much of a recency bias, but it’s probably still a cloud of terms.
Those who paid attention in history class might even recall some of these things:
Southern Democrats
Tammany Hall
Bull Moose Party
Missouri Compromise
Whigs
But, chances are, they probably only understand them insofar as they understand that “the mitochondria are the powerhouse of the cell”. Even figuring out where to place them in that word cloud may be a bit tricky. Conventional wisdom would say that placing them into the word cloud requires learning more facts, perhaps some (shudder…) dates. But that will only make the cloud bigger. A bigger accumulation of facts isn’t what you want. You want to see the structure of history.
The structure of history
The whole reason you find history so interesting is because you want to know the grand scheme. History is a four-dimensional machine constantly turning, the sum total of interactions between humans and their world over the course of millennia. Knowing how it works is seductive. You want to break open the machine and see the gears. So you read about history, you accumulate facts, and then you attempt to organize them into a cohesive picture, a glimpse into the guts of the machine.
But, no matter how much you read, you never seem to get there. With each new fact you learn, that cloud of concepts might resolve into a shape, or it might fuzz out, even more muddled than before. I’ve got bad news: If you continue trying to learn history by merely piling on more facts, it will only ever collapse back into “a bunch of stuff that happened” and the grand structure will elude you.
But I also have good news. There are those who have had a glimpse at the underlying machinery of history, those like Jared Diamond, who wrote Guns Germs and Steel, or David(s) Graeber and Wengrow, who wrote The Dawn of Everything. Through years dedicated to focused research with copious amounts of data, through the brute force of concentrating all their cognitive resources on the problem, they have seen The Structure in their minds.
These scholars publish books on their models, in an attempt to let us also see the structure. But there’s a problem. We cannot see what they see. Our view of the structure is blurrier. They came to see the structure through master-level knowledge of their domains. Their expertise allowed each individual piece of knowledge to take up less space in their minds, allowing them to focus their cognitive energy on putting them all together. They have seen it for themselves, but you are only imagining what they can see based on their description.
While reading about their models and theories, you will certainly have a head start on your way to truly seeing the structure of history, but your picture will not be as clear. Just reading the book itself will take time, but then you will have to do your own research. You will need to test out their model for yourself. You will need to compare it against your own knowledge, to contextualize it, to make it your own, and to form a picture of the structure in your mind.
But what if you could get there faster? What if, instead of telling you about the structure, they showed it to you? But how would they do that? How do you draw a four-dimensional machine?
Without maps, there is chaos
I already know you recognized each of the above shapes in the time it takes to blink.
The power of maps is obvious when you are trying to navigate in a city for which you’ve never seen a map. The sheer complexity of the world feels overwhelming until it has been abstracted into a clear gestalt. The map becomes your mental model. If you have never been to a city, it helps you learn your way around by giving a greater context to everything you see. If you have already been there, it strengthens your understanding of that city by weaving together the collage of mental snapshots from your personal experience. Just seeing the map once is sometimes enough.
But, as those silhouettes above might suggest, maps are not only useful for navigation. They have a deeper ability that most don’t realize. Try and conceive of all the things you know about the world, the countries, the cities, the languages, the religions, the currencies, the politicians, the landmarks without picturing the shapes of those places on a map. You can’t. Your brain has already linked all that information together. Images are not only a way of receiving information, but also of storing it.
We are a visual species. Our brains have a hard time with abstract concepts. This is why our societies create graphical symbols for nations, religions, and ideologies. Those symbols let the powerful “graphics cards” in our heads process large amounts of information, leaving our prefrontal cortices free to actually make insights about the things we know. On a map, each of the countries, represented by a shape, is a symbol. Each symbol holds everything you know about that country. And the spatial relationships between those symbols hold everything you know about the relationships between those countries.
In other words, would you be more able to understand the state of global human development by reading a table with the HDI score of every sub-national region on earth, or by looking at this map?
It’s a rhetorical question. Even understanding the world as it exists at this moment would be impossible for us without maps to communicate it visually. Our view of the world would be as hazy and cloud-like as our view of history currently is. So, we must apply the power of maps to history in order to have the kind of understanding that experts like Diamond, Graeber, and Wengrow have.
But history is not about “this moment”, nor is it about any other one moment in the past. History is the complex web of interactions between humans and their world continuously unfolding over time. Geography exists in three dimensions, history happens in four. How do you make a map of that?
Mapping in 4D
To start with, no, you can’t just show the same map at different points in time and call it a map of history. That’s not how human perception works. The human ability to compare the same spot on a map across several iterations is very limited, let alone the ability to compare multiple regions at once. Sequential maps are better than nothing at all, but they still atomize geographical information within each depicted time period, and atomization of information is how we end up with those clouds of facts.
To create a map of history, we will start with time. This is the easiest part. Time is a non-negotiable when discussing history, and there is an established way of depicting time in a graphic form. We can create an X axis, and track change over time with a line graph. Of course, with only one line, we can only depict a tiny slice of history. This is the percentage of votes that the Democratic party got in each election.
Line graphs can show us a shape of time. If the map’s ability to show us the world was a remarkable leap for human cognition, the line graph’s ability to let us “see time” would be considered almost black magic by the ancients. But seeing a single variable across time is not seeing history. History is the interaction of many variables, so it cannot be viewed one variable at a time.
In the case of the history of American presidential elections, we would need to track at least 50 variables, one for each state. Attempting to do this with a line graph would result in a gruesome mess, so we are going to replace the lines with color coding. Not only will that create a cleaner visual, but it lets us indicate the winning party in each state by using both hue and intensity. Now we are tracking at least 100 variables in this graph assuming at least two parties running in each election. This is the result:
Now we have a heat map. While it’s an improvement over the line graph in that it carries far more information, it still doesn’t count as a “map of history”. A map allows you to see the grand structure. By sorting the states alphabetically, it shatters the structure. It would be like taking a map of the 2004 election results and jumbling all the states randomly. The entire “Jesusland” concept would be lost.
To make this table of states and years into a “map of time”, we have to assemble the 4D machine that is history. We have over 100 variables in our heat map representing states and votes, but we cannot make out “the complex web of interactions” between them that would make this a map of history. For that, we need to do something which will bring those interactions to the surface.
Those interactions are present in the above heat map, but they are scrambled and all but impossible to perceive. To unscramble them, we need to change the ordering of the states on the Y axis. Just as the Jesusland map has a grand theme that only reveals itself when the states are in their proper places, our map of history can only reveal its patterns when it is properly ordered.
But what is “properly ordered”? The simplest interpretation would be to lay out the states on our axis so that states that are closer on a map would be closer on the line, and neighboring states would be adjacent on the line. This makes sense because neighboring states influence each other and are influenced by the same higher-order phenomena, and thus they would be more likely to have similar voting patterns, and that would in turn reduce noise and help resolve the “web of interactions”.
The obvious problem with this approach, though, is that it won’t always work because any state can have multiple neighbors on a map, but only two neighbors on a line. There are several possible ways of ordering the line. Some make more sense than others. Look at the example below of the six westernmost states, and decide which makes the most sense to you.
Here is where I make this article worth your time, and explain what a historigraph is. Since we cannot rely solely on geographical position to lay out the states, we need more variables. Remember that we are mapping history, not geography, so the spatial locations of each state are only part of the equation.
What that means is that the placement of each state on the axis will take into account not just its position a map, but its behavioral similarities to other states. As you no doubt know, the West Coast states and the Inland West states have very different voting patterns, so we need not even place them near each other at all. In fact, we could put them on opposite ends of the axis. If we continue laying out all the states so that the placement of any two states is based on a composite of their geographical location and their political and cultural commonality, we end up with this:
The above map is a depiction of how the 2D map is converted into a 1D line. It is shaded so that the states that appear near the top of the Y axis are darker and the ones near the bottom are lighter. This means that, anywhere there is a sharp break between dark and light (i.e. two states far apart on the 1D line), there is a significant cultural and political divide between the two. The sharpest divide between neighboring states in America is likely the one between the West Coast and Inland West, and it is reflected in the black-and-white contrast. Meanwhile, everything else in the country is a gradation.
The final product: Historigraph
You’ve already seen it, but here it is again:
The more you look at it, the more you will see that it reads like a map. Concepts that span both space and time are shown as “landmasses”. The deep blue “sea” in the middle is the Solid South, the long-time political affiliation of the Democrats and former Confederate states. The polarization of American megaregions “Jesusland” and the “United States of Canada” is shown clearly as a sorting of blue and red on the right side. And the chaotic mess of colors toward the left represents the decades-long struggle to form a stable opposition party to the Democrats.
To demonstrate that this ordering scheme wasn’t a fluke, or tailored to fit the data, here it is applied to a heat map of COVID cases in 2020. Note how you can actually see the pandemic radiate outward from New York, both to the north and south.
A guided tour of this historigraph
Not only does the historigraph allow you to read history like a map, but it also serves as a form of timeline. Unlike a regular timeline, which is little better than plain text and cannot be read at the macro level, the historigraph allows you to create your own timelines, like planning a road trip on a map. The various “coastlines” and “islands” on the map serve as waypoints. Once you know what each of them is, you can see the implicit timeline despite the lack of labels, just as you can identify Africa or Hawaii or Greenland.
What follows are two guided tours of the landmarks in this historigraph. A lot of what you read, maybe even all of it, will be stuff you already know. But being able to visualize all of these historical moments and periods as places on a map of time and space will allow you to “know” them in a whole new way.
History of American party systems
This is another one of those terms floating around in your head. A lot of people know that “the Democrats used to be the party of slavery”, but that’s about all they know. In fact, America’s two party system has had several iterations. The Democrats were originally known as Democratic Republicans until Andrew Jackson reformed the party in his own image, and we didn’t even have the GOP until 1854. Most people only know the Whig party for giving us a few forgettable presidents who died in office and gave us even more forgettable vice presidents.
Even the concept of monolithically “conservative Republicans” and “liberal Democrats” is a fairly new one, having only become true during the Reagan era. Given that American politics appear to be undergoing a radical shift in which the more extreme elements of the Democratic and Republican parties appear to be fragmented away from the centrists in either, we may well be witnessing the dawn of a Seventh Party System. If you want to understand how the dynamics of our two-party system works, read this one.
Political history of American regions
If you have read Albion’s Seed or American Nations, the concept of American cultural regions won’t be a surprise to you. Really, pretty much everyone knows that “the North” and “the South” are different, but those simplistic divisions fail to capture what American regions are really about. European settlement in America was not a uniform migration from England. Rather, at least four separate regions of Britain all sent a different demographic of their population to different parts of America for very different reasons. They were followed by Germans and Dutch, and each of these founding populations created cultures that still endure to this day. In a sense America is still composed of separate nations as different from each other as Italy and France.
This walkthrough highlights how these various regions continued and still continue to assert their identities through their voting patterns. The idea of “Red America” and “Blue America” may be a temporary illusion created by very specific conditions. I do highly recommend reading those books to learn more.
Once you have read through these, your knowledge of history should now feel a lot more coherent. All those facts you know now suddenly have a shape. Now you can start to think about history at a higher level because your brain doesn’t have to do all that work holding together those facts; the historigraph does that work for you. That was the point. This is just the beginning though. There are plenty of other types of data we can track and plenty of other times and places we can map.
Some predecessors I didn’t know about until recently
I had found myself wondering why nobody else had thought of this, because the concept seems obvious to me. The need for historigraphs should be clear, but the cause hasn’t been taken up. In the process of writing this article, I found a sort of prototype of the historigraph from 1769. I’m not sure if I am more astonished that it exists, or that it wasn’t the beginning of something more.
While the limited color palette and arbitrary Y axis (the placements of Africa, America, and Russia are inexplicable) hinder its readability, it’s clear that the creator, Joseph Priestley, was onto something big. While he made a few other proto-historigraphs like this, none were quite as ambitious, and nobody took up the mantle after him.
The only remotely equivalent thing that has been produced in the post-industrial era is this 1931 chart from Rand-McNally that they call a “Histomap”. It was actually this one which inspired me to create the historigraph.
While it’s certainly fascinating, it lacks the standardized grid of Priestley’s graphic and it’s hard as hell to read. It’s barely better than a timeline when it comes to macro readings. The historigraph concept seems to have died out before it was even born. Let’s hope the same doesn’t happen this time around.
If you found this useful, please spread the word
There is a lot more work to be done exploring the possibilities of historigraphs. I have wanted to make examples with much larger data sets, both in terms of time and space. Viewing American election results by county would be even more interesting, but getting the data isn’t as easy as you think, even in this day and age, and then creating a scheme to order all the counties will be much harder than the states, and it will need some math and more data. Plus, I’d like to experiment with a historigraph that doesn’t use space for the Y axis, but rather other non-linear information. It’s only worth doing if I know people want it. So you know what to do.
The trash collected by the eyes and dumped into the brain, said it tears into our conscious thoughts. You tell me who's to blame.