We are now a highly divided country, but we have had divisions that have affected politics, government, and our society since the inception of the nation. For over seventy years, we had slave states versus free states, which morphed into South versus North. We have had free trade versus protectionism; the gold standard versus silver; imbibers versus teetotalers; labor versus management; men versus women; everyone versus plutocrats; Protestants versus Catholics; everyone versus Jews; blacks versus whites; Italians versus the Irish; suburbs versus cities; urban versus rural; war proponents versus war opponents (and on a different front, everyone versus the New York Yankees; everyone versus Tom Brady fans; everyone versus Ted Cruz; and almost everyone versus that My Pillow guy).

 We tend to believe that our politics reflects our divisions, but it does more than that. Our politics helps to create the divisions and does less than ever before to bridge them.

The two major parties are increasingly ideological or, perhaps more accurately, partisan. When was the last time that there was a conservative Democrat or a liberal Republican? There are many reasons for this. The successful Republican “southern strategy” killed the old Southern Democratic party and thereby most conservative Democrats. Newt Gingrich and Tom Delay made Republicans, in effect, swear loyalty oaths to one brand of Republicanism that excluded liberal Republicans. Gerrymandering has produced more safe seats in both state and national legislatures creating ing districts where officeholders and seekers do not have to appeal across divides for votes. Robert G. Kaiser in Act of Congress: How America’s Essential Institution Works, and How It Doesn’t points out that as cooperation between the parties has declined, the demonization of opponents has become more common which further increases divisions. (Kaiser says this trend has been aided because members of Congress no longer spend uninterrupted weeks and months in Washington getting to know each other. Instead, with modern transportation, they have three or four-day Washington work weeks and then go back to their home districts.) Trump is not the creator of this path of demonization and divisiveness; he is merely an extreme example of its danger. He and other politicians seek to divide with the hope that their cries of the devil will energize their supporters who will turn out to vote more than their opponents.

Besides the major chasms that the parties have fostered, however, politics today creates many mini-fractures in the societal landscape that have been deepened by modern media. Jill Lepore suggests in These Truths: A History of the United States that this started with the 1960 presidential election when the Democrats turned to “data science” and hired the Simulmatics Corporation. (Jill Lepore, who seems to publish something interesting at least once a month, has a new 432-page book out about Simulmatics, If Then: How the Simulmatics Corporation Invented the Future. I have not [yet?] read it.) That company got punch cards from pollsters of recent elections, fed them into an early computer, and “sorted voters into 480 possible types.” One category, for example, was “Eastern, metropolitan, lower-income white, Catholic, female Democrats.” Another was “Border state, rural, upper-income, white, Protestant, male, Independents.” Such data dredging revealed shifts in voting patterns not apparent before. So, for example, the Democrats discovered that between 1954 and 1956 a small but significant shift by northern Blacks to the Republicans had occurred in eight key states. This helped to propel the Democrats to put a civil rights plank into their 1960 platform and encouraged John F. Kennedy, who had showed little interest as a Senator about the issue, into supporting civil rights.

We continue to see similar data today. Politicians seem to assume that on many major issues—abortion and gun control, for example—minds are made up, and voters and can’t be persuaded away from the candidate they already support. Accordingly, elections are seldom analyzed along ideological lines. Instead, analysts and strategists turn to the 1960 approach but with increasingly sophisticated demographics. “Male whites without college degrees and over fifty living in a rural area;” “white suburban women with children who work outside the home with family income above $75,000 per year” are now categories that are refined and then refined some more.

Of course, this in some way is just an extension of what has been going on since the Kennedy-Nixon election, but now there is a difference. However it was that JFK decided to be a civil rights candidate, this stance was apparent to all Americans. He was basically the same candidate in every region and with every demographic group.

JFK, however, did not have the tools available today. Through social media, the internet and search-engine advertising, increasingly smaller demographic groups can be targeted, and a distinct message can be tailored for each groiup without that campaign reaching all Americans. The candidate I may think I know might appear to be different from the same candidate that you think you know. We may vote for the same person, but we may be voting in essence for different candidates, and a little wedge is driven between us as a result. Instead of broad expanses of shared perspectives, we have a many-fissured landscape that looks different from every perspective.

Our current political process does not just exploit the divisions among us; it helps to create them.