The First Party System
This is the first installment of the “America’s Historical Party Systems” series. It’s advised to read my piece on CIZST before starting here. And stay tuned for more!
Welcome, friends! This article took a lot longer than I was expecting; the core research was pretty time-consuming. But I think at least it sets a nice expectation for the future: about 2 months per party system. At this rate, we’ll be done with all 6-7 by 2023! Anyways, let’s dive in.
Introduction
The First Party System, officially 1796-18281, presents a difficult data set for analysis. Before 1796, there weren’t any formal parties, but Congress had already separated into Pro- and Anti-Administration factions, which later became the Federalist (F) and Democratic-Republican (DR) parties, respectively, with the “Pro” side managing slight majorities up to 17962. These two parties held steady until the aftermath of the War of 1812, when the Federalists imploded, leaving the DRs with a national hegemony that itself quickly imploded under the pressure of sectionalism3. A four-way race in 1824, all under the DR label, erupted into a party split that quickly re-consolidated into two coalitions by the next presidential election in 1828: the (Jacksonian) Democrats and National Republicans (NR)4. Most of the pre-Era-of-Good-Feelings DRs became Democrats, and most places where Federalists had formerly ruled became NR strongholds.
So, for some of the analysis, I had to ignore Washington’s two terms, as well as the Pro- and Anti-Administration labels. I don’t consider this a major loss; they fall before Wikipedia’s official start of the First Party System, and we’re also not losing much information there in terms of the dynamic nature of the period. Washington’s two terms were mostly a period where everyone was too afraid to actually declare as partisans just yet, but kept getting polarized over every political fight that arose, motivating them to set up factional newspapers, develop wedge issues, and other precursors to overtly partisan institutions.
Systemically, the major trend was towards zero-sum in the presidential election, and away from zero-sum in the House. The 12th Amendment was adopted in response to the failures of the 1796 and 1800 elections; while it reaffirmed the commitment to the need for an absolute majority of electoral votes (making it a consolidation of zero-sum incentives in presidential system), its immediate aim was to produce unified presidential/vice-presidential administrations. At any rate, developments at the state level were more dynamic, contentious, and interesting. The overall trends toward zero-sum Electoral College delegations and apportioned House delegations obscure the more trenchant observation that states generally did not change their systems without an impetus. These fell into three categories: power-grabs, power-sharing deals, and the Jacksonian groundswell of support for House apportionment. It shouldn’t be a secret that I suspect that democratic experimentation in the early republic was decidedly motivated more by political opportunism and entrenched interests than high-minded idealism, but I do hope to see what the data and history show us.
Since we’re framing this all within CIZST, it demands that we examine what the relevant intersections of identity and interest were on the ground. The Eleven Nations model gives us a good starting point: the Tidewater, Deep South, Midlands, New Amsterdam, Yankeedom, Greater Appalachia, First Nations, and New France were all present in the fledgling United States. Within the American electorate, we have a deeper dive into ethnic demographics: the English groups of Scots-Irish, Norman Cavaliers, New Englanders (Puritans and Quakers, mainly), and the South’s Barbadian5 slave lords. Among the non-English, there were free Blacks, a prominent Dutch population, a growing minority of German immigrants, other New World colonists descended from France and Spain, and tiny minorities of other recent European immigrants, the largest perhaps being Italians. Religiously, each of the English ethnicities was mostly tied to a prominent Protestant church, the Germans were split between Catholic and Protestant, and the French, Spanish, and others were mostly Catholic. Finally, in terms of regional interests, the North was in the early stages of industrialization, eyeing banking and trade infrastructure to establish competitive footholds in the global market, while the South was mostly interested in (1) trade protectionism and (2) a slow but growing interest in expansionism.
The Data
Partisan Control
Let’s start with just a simply “party control” breakdown: what were the partisan outcomes of each election, in terms of control of the Presidency, House, and Senate?
For these next ones, please forgive the discontinuities and odd shapes in the plots around 1822-1824. If I were doing this in MATLAB, best believe these would be immaculate, but this is just in Excel, and not really worth the effort to make them as neat as the above chart from Wikipedia.
What we see here is no mystery. The Federalists started off with somewhat of a fighting chance, but overplayed their hand with the Alien and Sedition Acts, provoking a backlash that they never recovered from. We do see a bump for them after 1812, but it’s mostly attributable to redistricting from the 1810 Census adding seats to their strongholds. Although this points to the fact that Northern population growth was systemically underrepresented, the political truth was that the tide had already turned against the Federalists — as we’ll shortly see, the North was too internally divided to remain united behind the Federalists, while the South was almost completely united from the start. The Federalists were also on the wrong side of the War of 1812; although they were seen as weak on Britain, and mustered a strong opposition to the war, they snatched defeat from the jaws of victory by igniting a backlash against revelations about the secret proceedings at the 1814 Hartford Convention, where they aired a laundry list of grievances over the run-up to the war, and openly pondered secession if their rather extreme demands weren’t met.
Sectional Unity
The point about sectional unity segues nicely into our next topic. One way of looking at it is to see which states swung more frequently than others, among all branches.
Right off the bat, we see Delaware, Maryland, Massachusetts, and New Hampshire among the biggest swingers. New York, New Jersey, Rhode Island, and Vermont form something of a middle class, with Connecticut bringing up the rear.
Where this really pops is if we regroup the states by region — using the 11-nation model as a base, let’s use “The South” for the Tidewater plus Deep South, split the Western states off from the Midlands, and combine Yankeedom and New Amsterdam into “The North”.
The West hadn’t been around long enough to have many swings, but what’s remarkable is just how much more the North and Midlands swung than the South, given just as much time.
Zero-Sum and Electoral Evolution
Now here’s the real meat of our CIZST analysis. One thing that’s unique about the First Party System Era over all the others we’ll study is that the methods of election were highly varied for the House and Electoral College.
We don’t need to go into all the intricacies of the different electoral systems in use at the time, but for the House and Electoral College, they broadly fell into two categories: “apportioned” and “zero-sum”. Although apportionment mostly meant single-member districts, which we at the Discourse often (rightly) deride as zero-sum in today’s context, they were as progressive as it got during the First Party System. By contrast, At-Large was a particular favorite of the time for dominant state parties to monopolize their state’s federal power in zero-sum fashion, while pretending to love democracy and nonpartisanship6.
Right off the bat, what we see is that zero-sum was highly-correlated to single-party domination of a state’s House delegates: from 1788-1828, 85% of non-small states had single-party delegations, and the number rises to 95% for the small states7. By contrast, apportioned non-small states only ended up with single-party delegations 26% of the time8. In the Electoral College, only 21 of 174 (12%) state elector slates were ever split. Of those 21, only 6 occurred under zero-sum systems. Perhaps a better way to frame it, though, is that apportioned states were better at achieving split outcomes: 15 out of 42 (37%) elector slates, versus 6 out of 132 (<5%) for zero-sum states. Apportionment was not perfect proportionalism, but it was relatively better than the tyranny of at-large single-party domination.
However, the broader question we’re trying to ask here is about evolution. State-by-state analyses will have to wait for later deep-dives9, but there are several broad trends that we can see from the following charts on electoral methods.
Note: “Legislature bullshit” generally means at-large methods that are essentially zero-sum.
The oldest and newest states tended to be more zero-sum, but likely for different reasons. The former would have suffered more domination by entrenched interests. And the latter typically lacked the population to merit anything more complex.
States dominated by single (Woodard-ian) nations tended to be more zero-sum.
Northern states split between multiple nations tended to be apportioned. The more nations, the more complex the apportionment — essentially, a form of negotiated power-sharing.
The South generally adhered to apportionment, except in new states which didn’t have enough population to justify it.
Newer states were more willing to adopt the modern arrangement of “Winner-Take-All (EC) + Apportionment (House)”.
Some notable characters and experiments pop out. Connecticut and Delaware were each outliers within their own regions for being dominated by conservative attitudes and entrenched parochial interests. Maryland, New York, and Pennsylvania saw durable multi-member district arrangements, but Massachusetts and New Jersey’s experiments each only lasted a single cycle. Vermont likewise appears to have had some turmoil in the 1810s. Georgia had a single-election experiment with apportionment in 1826 — the height of the Jacksonian upheaval. And the South, at which an inveterately biased Northerner such as myself would love to cast every aspersion of antidemocratic subversion, was a bastion of apportionment, which we’ve noted was about as progressive as electoral systems got.
Odds & Ends
Since we’re nearing the email length limit already, some stuff has been triaged, but there are some minor thoughts I wanted to bring up that will hopefully spur discussion going forward.
The franchise was highly restricted. Turnout ranged in the thousands to tens of thousands, not millions. It’d also be interesting to see if turnout was affected by the lack of established party brands, but unfortunately there just isn’t any reliable data from the era.
Even though we’re still talking about a highly undemocratic era, electoral outcomes are still informative from a CIZST perspective because they tell us how the people who did have power felt and acted.
As I spin up work on the Second Party System, I’ll also be doing a series of shorter state-level deep dives on how the electoral systems evolved in the states I identified as “notable characters and experiments”.
According to Wikipedia. Remember, we’re using Wikipedia’s definitions in order to avoid getting drawn into silly academic debates.
For anyone who cares, Washington basically coded as a Federalist, which is why the Pro-Administration faction became Federalists.
This is academic-speak for “regionalism”, but is meant to imply a larger and more definite scale than “regionalism” might. For instance, 1824 actually saw two different candidates who were popular in what we might call “the South”. Sectionalism allows us to consider the differences and commonalities that might lead to a temporary fracture within an otherwise cohesive region.
Note: Not the same as the GOP/Republicans. They quickly abandoned the NR label to become the Whigs.
This descriptor from Woodard is notable because of the special mix of brutal slavery fueled by cronyism and speculation that arose in the Caribbean and was imported to the Deep South. It’s hard for a non-historian like myself to pin this down to a specific British Isles ethnic group as the sole originator, and such an originator would actually obscure the most relevant fact: that the founders of the Deep South were hard men who were used to cutthroat speculation, under the auspices of a distant crown but often simply at the barrel of the gun of whoever managed to come out as “top dog”.
One caveat is that new states typically started out with only one district, which applied to 10 delegations out of 87 over the period.
This matters because we don’t want to dunk too much on small states, especially if they only have a single seat, but it still helps our point that when we ratchet up the zero-sum-ness, we get more single-party domination.
I’d love to see how this stacks up to later party systems, but that’s for later analyses.
We’re reaching the email length limit, so some stuff is going to have to get triaged!