“Count every legal vote” sounds like a good, rule-abiding idea.

Too bad it’s also racist.

Even before Donald Trump lost, he and his supporters settled on “count every legal vote” as their mantra — at least in cities like Philadelphia, Detroit, and Atlanta, all of which have a plurality if not a majority of Black residents. It’s a crucial difference from “count every vote.”

Adding the adjective legal implies the presence of illegal votes, which lawsuits, the Department of Justice, and even super-sleuth Rudy Giuliani have been unable to provide evidence of. (Maybe because he’s been looking in Holmesburg.) But that doesn’t matter to Trump, because evidence isn’t the point here. It’s the insinuation of illegality in service of eliminating Black votes.

News media use the descriptor Black three times as much as white, which normalizes white and others Black. Similarly, legal vote others ballots from areas that aren’t predominantly white.

The term legal vote has a history that’s fraught with racial complexity.

In America, the term legal vote has twice peaked in usage: 1811 and 1877. Though we can’t say for sure what caused those spikes, two historical moments stand out: 1810 marked the start of voting restrictions in many U.S. states, including Pennsylvania, Tennessee, Maryland, New York, and North Carolina, where free African Americans lost the right to vote. And 1877 followed the acrimonious election of 1876, when Democrat Samuel Tilden won the popular vote over Republican Rutherford B. Hayes by three percentage points. But four states' results were disputed, so neither candidate had an Electoral College majority. (Sound familiar?) The resulting Compromise of 1877 gave Hayes the presidency as long as he withdrew federal troops from the South and ended Reconstruction. Black and Southern Republicans called this the Great Betrayal, and it led directly to poll taxes, literacy tests, and other tools to disenfranchise Blacks across the South for decades.

(The third, much smaller peak followed the contested 2000 election, in which the Supreme Court ruled that “a ‘legal vote’ … is one in which there is a clear indication of the intent of the voter.” But that was in the era of dimples and hanging chads, ballot types that are blessedly irrelevant in 2020.)

The racist usage of legal and illegal is clearest in the context of immigration. Particularly when used instead of the more factual undocumented, the term illegal — especially when it’s a noun — is both highly offensive and all too common. On the Offensive, a new book on language by Karen Stollznow, notes that a University of Oxford study found that in writing, illegal is the most common modifier of the word immigrants, despite widespread agreement that the term is both harmful and disproportionately used to describe nonwhite immigrants.

The most pernicious usage of legal votes is its insinuation that legal votes come from white, rural areas and illegal votes come from Black, urban areas — like Philadelphia. Like the 2017 Supreme Court case in which a federal appeals court struck down North Carolina’s voter ID laws for “target[ing] African Americans with almost surgical precision,” Black Americans' right to vote has always been under assault, and the language that enables that assault is often coded. Implying that legal always equals legitimate is the code du jour: It ignores the surgical precision of such laws that have disenfranchised voters, from poll taxes to literacy tests to grandfather clauses. As in grammar, blindly following rules without knowing the reasoning behind them can make you look foolish, and ignores that some laws are passed with nefarious intent.

This racist history allows those who control the narrative to determine which votes are legal. When one candidate can choose whose votes count, why bother having an election at all?

The Angry Grammarian, otherwise known as Jeffrey Barg, looks at how language, grammar, and punctuation shape our world, and appears biweekly. Send comments, questions, and tri-chads to jeff@theangrygrammarian.com.

Read more from The Angry Grammarian