Some ty­pog­ra­phy con­cepts have evolved in re­sponse to changes in tech­nol­ogy. For in­stance, in the dig­i­tal age, a car­riage re­turn gets in­serted at the end of a para­graph, whereas on a type­writer it was used at the end of a line. And to­day, the word font has largely ab­sorbed the tra­di­tional mean­ing of type­face.

But other con­cepts have been sur­pris­ingly durable, like the em. In hy­phens and dashes, I men­tioned that the word em de­notes a ty­pog­ra­pher’s mea­sure­ment, not the let­ter M. The em size of a font is the same as its point size. This was true for hun­dreds of years of metal type.

It’s re­mained true in the dig­i­tal era. To­day, the em is im­ple­mented in soft­ware rather than metal. But it still rep­re­sents the same thing: the max­i­mum ver­ti­cal size of the let­ters in the font. Then and now, two fonts set at the same point size will ap­pear to be dif­fer­ent sizes if one oc­cu­pies less space on its em.

“But why not just make the point size equal to some­thing spe­cific, like the height of the cap­i­tal H?” Not a bad idea, but since fonts vary widely in their de­sign, there’s no char­ac­ter you could pick that would al­ways be con­sis­tent. Nor can you sim­ply scale every­thing to fit the em—it’s clear from the il­lus­tra­tion above that if the cap­i­tal H oc­cu­pied the whole em, de­scend­ing let­ters (like g j p q y) wouldn’t fit.

In­stead, by re­ly­ing on max­i­mum size, the em-siz­ing sys­tem can be ap­plied to any font. This ends up be­ing a use­ful con­ven­tion for au­to­mated type­set­ting. But as you learned in the pre­vi­ous sec­tion, it’s also why you can’t de­ter­mine the point size of a font by mea­sur­ing it directly.

If you think that’s merely nerdy trivia, you’d be wrong. A 2012 case in the Michi­gan Supreme Court turned largely on the mean­ing of point size. Michi­gan state law re­quires cer­tain bal­lot mea­sures to be “printed in cap­i­tal let­ters in 14-point bold­faced type”. One cit­i­zen’s group had col­lected enough sig­na­tures to put their mea­sure on the bal­lot, set­ting their font to 14 point in the usual way. But a sec­ond group chal­lenged the va­lid­ity of the mea­sure, ar­gu­ing that the 14-point re­quire­ment ap­plied to the size of the “cap­i­tal let­ters”, not the “type”, and thus the bal­lot mea­sure had not been printed large enough.

As a ty­pog­ra­pher, it was clear that this case should’ve been re­solved swiftly. Point size has had a con­sis­tent mean­ing for hun­dreds of years. It’s never had any­thing to do with the size of cap­i­tal letters.

More­over, as a law­yer, it was clear that if this fact weren’t au­to­mat­i­cally clear from the plain mean­ing, it could’ve been es­tab­lished by ci­ta­tion to au­thor­ity (this book is just one of many op­tions), ex­pert tes­ti­mony, or even good old ju­di­cial no­tice. What was left to ar­gue about?

A lot, ap­par­ently. The ap­peals court ruled that “14 point” man­dated the size of the cap­i­tal let­ters, not the over­all type. For­tu­nately the Michi­gan Supreme Court re­versed this de­ci­sion—adopt­ing the cor­rect de­f­i­n­i­tion of point size—though the vote was alarm­ingly close.

Why alarm­ing? None of the jus­tices seemed alert to the po­ten­tial con­se­quences of Michi­gan’s high­est court re­defin­ing point size as a mat­ter of law. Had it done so, point size would have meant some­thing dif­fer­ent in Michi­gan than it does else­where. It would have been like the court say­ing “π is such a long num­ber—we hereby or­der it to be 3.”