No Roman was ever called Caius

I have been reading Robert Harris’ Cicero trilogy with great enjoyment. However, one thing jars. Men whose praenomen (“first name”) was Gaius are routinely called “Caius”. For example, in Lustrum (the book that I’m currently reading), Cicero’s consular colleague in 63 BC is named “Caius Antonius Hybrida”.
Of course, this problem does not belong exclusively to Harris. Other authors are in the habit of rendering the praenomen Gaius as “Caius”. The reason is a simple misunderstanding.

The Latin language derived from Etruscan, which had no letter B, D, or G. For example, the word magistratus (“magistrate”) is rendered as macistratus on early Roman inscriptions (e.g. CIL IX, 782 = Dessau, ILS 4912; or CIL VI, 1300 = ILS 65, pictured below — note also “Cartacinienses” for the people of Carthage on line 9).

Consequently, the praenomen Gaius had to be represented as “Caius” and was abbreviated with the letter C. This rule persisted down through the ages, so that, even with the introduction of the Latin alphabet (with its B, D, and G), the letter C was retained as the abbreviation for Gaius. Similarly, Cn. remained the abbreviation for Gnaeus.

The problem has perhaps been exacerbated by the epigraphic convention that calls for the expansion of the abbreviation C. in Roman inscriptions as “C(aius)”, instead of pointing out that it simply stands for the praenomen Gaius.

In fact, the historian A.R. Birley once suggested that epigraphers should employ the convention “C. (Gaius)” to indicate this (L’Antiquité classique Vol. 70, 2001). He called for the banishment of “Caius” and “Cnaeus”, and pointed out, quite correctly, that the Greek writers of the Roman era employed the word Γαίος (not Καίος) for Gaius (thus indicating the correct pronunciation).

So let us have no more Caiuses and Cnaeuses! No Roman was ever called by those names.