The serifs have appeared for the first time in the Latin inscriptions cut in stone.

So if there was any technological reason for them, it was either due to the stone carving techniques, or more likely to the limitations of writing with a wide brush, because the stone was carved after the letters had been painted with a wide brush, to guide the carving.

After the Imperial Rome, the serifs have continued to be used when writing on different supports mainly due to tradition.

The serifs are not the only difference between serif and sans-serif typefaces, the second difference is that the sans-serif letters are drawn with lines of uniform width, while the serif letters are drawn with a combination of thick lines and thin lines.

There are some modern sans-serif typefaces, like Optima and many others inspired by it, which are intermediate between classic serif and sans-serif typefaces, by lacking serifs but using variable-width lines, like the serif typefaces.

Even if the serif fonts are the traditional fonts, used almost exclusively until the 19th century, when the sans-serif fonts became popular for certain uses (e.g. for titles, for advertising or for newspapers printed on cheap paper) there are good reasons to use them besides the tradition.

Their 2 extra features, serifs and variable-width lines, make the letters more distinctive, less similar to each other, and in the opinion of many people, more beautiful.

Because of that, when rendered at very high resolutions, most people consider the serif typefaces more legible, even if there are also younger people, who have read few books on paper, but who have been accustomed with reading sans-serif typefaces on low-resolution displays, so they may prefer the sans-serif typefaces.

As another poster has already said, the advantages of serif typefaces are achieved only at high resolutions, i.e. preferably on 4k displays or better and when not using stupid dpi scaling, but actually using the high resolution for rendering. On the many garbage laptop displays with low resolutions, the sans-serif typefaces are almost always better.

I want to add that why I am not completely convinced that it is better for the letters to have serifs, I definitely believe that the letters drawn with constant-width lines are much uglier than the letters drawn with variable-width lines.

Because of that, even if I use frequently sans-serif typefaces, I use only non-traditional sans-serif typefaces with variable-width lines, e.g. Optima Nova, Palatino Sans or Trajan Sans.

However, such typefaces with variable-width lines also need high-resolution displays to be useful. On displays with less than 4k resolution, their appearance at normal text sizes is degraded.

So would you then say that the popularity of sans-serifs as body text on printed media such as posters, back-covers, consumer packages, etc. is due to them preferring style over legibility?

It's an interesting question. I know that ink bleed was taken into account in type design, which actually led to some early digital versions of classic typefaces being poor renditions because they were too literally following the cuts of the metal and ended up being too thin.

The traditional rationale I come across is that serifs "lead the eye," but I have no idea if there's any real research about that. There's probably some truth to broadly stereotyping sans-serif fonts as generally more legible and serif fonts as generally more readable -- but I'm sure there's a myriad of exceptions. (Like, uh, Myriad, a sans-serif explicitly designed to be good for body text.)