Many people understand, at least on some level, that gender and family structure play pervasive roles in American politics. Political scientists have documented, for example, how women tend to vote for more progressive policies and candidates than men do, and how marriage tends to correlate with more conservative preferences among both genders. And of course some family effects go beyond gender, such as the electoral boost of coming from a brand-name political family.
But the shape of American politics might owe far more to gender and family structure than is broadly realized. In our research into the Constitutional Convention of 1787, we recently discovered that these factors might have molded the institutional design laid out in the Constitution itself: The more sons a Founding Father had, the more that Founder supported a stronger, more centralized federal government; the more daughters, the less he did so.
It’s no secret that when the U.S. Constitution was first drafted, only men were in the room. That undoubtedly affected the document’s final form, but, though we might wonder how things might have gone had women occupied seats at the Philadelphia convention, speculation is the best we can do.
We do, however, have information on another way that gender seems to have mattered to the delegates at the convention: whether they had sons or daughters.
In the modern context, political scientists have observed that legislators and judges with more daughters tend to vote and rule in favor of women’s issues significantly more than their counterparts with more sons. We have no reason to believe that those sorts of effects are limited to the present day. The politics of the late 18th century were far different from ours, to be sure, but also far more defined by the rigid gender norms of the period.
As a result, political elites like the Framers would necessarily have had sharply divergent expectations for their children. Sons, by convention, inherited virtually all of the economic and political power of their fathers. These sons would have been expected to someday go on to hold public offices and be in other positions of authority. Daughters, by contrast, would have been expected to marry and remain at home to raise children. They might have gone on to exercise some (occasionally powerful) influence through their social networks, but that influence would have been informal and local. No American woman held federal office until Jeannette Rankin of Montana was elected to Congress in 1916.
Delegates brought these expectations with them to Philadelphia. And, as documented in the convention’s records, they at times wondered out loud about how their children would fit into the fledgling republic they were building. In a debate over the equal footing of future western states, Roger Sherman of Connecticut observed that “we are providing for our posterity, for our children & our grand Children, who would be as likely to be citizens of new Western States, as of the old States.” Elsewhere, Gouverneur Morris of Pennsylvania and Elbridge Gerry of Massachusetts both specifically noted how powerful fathers would seek a place in government for their sons—daughters got nary a mention.
But did those ideas affect how delegates voted? We looked to see if any relationship existed between how an individual voted and whether that delegate’s children were sons or daughters (a random variable, making this a natural experiment). We found that the more sons a delegate had, the more likely he was to vote for measures that would centralize and strengthen the federal government’s power. The opposite was true for those who had daughters.
The effect not only was statistically significant but had as large of an impact on voting patterns as any that political scientists have measured empirically before. Each additional son was associated with an 8 percent increase in the likelihood that a delegate would vote for a pro-federal-government measure. Each additional daughter was associated with a 5.5 percent decrease in that likelihood.
What could explain the voting differences between fathers of sons and fathers of daughters? A few explanations are possible.
First, fathers of sons might have seen the convention as an opportunity to expand their family’s power and influence. Though it isn’t often thought about in these terms, the U.S. Constitution was, in at least some senses, the largest single expansion of government authority in American history. The delegates with sons could be quite confident that their family would share in that new power.
There might be a related but less cynical possibility. Because fathers of sons knew that their children would occupy the new offices, they might have been more likely to trust the federal government with greater power. The fathers knew their sons and believed their grown children could wield power responsibly.
Finally, fathers of daughters—who had neither of those two incentives—might have preferred to maintain the status quo. The Articles of Confederation kept power local, where daughters might have had some influence, even if through informal channels. Delegates with daughters might not have wanted to stray too far from that model.
Whatever the reasons, the delegates’ children had powerful effects on how they voted—and therefore on how the Constitution took shape and, ultimately, on how our government looks today. Many votes at the convention were close. A different combination of sons and daughters could have flipped enough state delegations to alter the document and shift the balance of power toward the states or the federal government.
For as much as Americans talk about the Founding Fathers, we talk shockingly little about the Founders as fathers. That is a mistake. Family relationships and their gender dynamics played a key role in setting the terms of political engagement established by the Constitution, just as they pervade our politics today.