Roman type has serifs, whereas Gothic type
does not. It is often assumed that the serifs make Roman type more legible, but this a
dangerous generalization to make. A lot of research has been done in hopes of answering the
question whether the assumption is correct. However, no clear answers have been found. The
problem is that each study provided a comparison of a limited number of typefaces, resulting
in various outcomes. Among others, Phillips (1977), Brachfeld (1964),
Crossland and Johnson (1928), Pyke (1926) and Kerr (1926), agree that serif typefaces are more legible. Tinker and Paterson (1928) have not found statistically significant
differences, while Ovink (1938) found some
Gothics to be more legible than some Romans when read from a distance. Watts and Nisbett
(1974) claim that Gothic type is more legible, citing the findings of Burt (1959) and Prince (1967). Zachrisson (1965)
claims that the absence of serifs may make a letter less defined, but that this does not
necessarily mean that the type has lower legibility. He argues that people were less familiar
with Gothics in the 1930’s, when much of the comparative legibility research was done.
Consequently, subjects had less reading experience with Gothic type than with Roman type. Moreover, a reader who sees a newly introduced typeface for
the first time could conclude that it has been imported from elsewhere and should thus be
mistrusted (in the 1930’s); unfamiliar becomes unwanted (chapter
4). Hartley and Rooum (1983) found
results of research in this area to be non-conclusive.
As long as research produces contradicting results, it can not be assumed that serifs
guarantee higher legibility. Interest in the possible benifits of serifs seems to have become less, since the notion has become
widespread that differences in levels of legibility between typefaces are too small to have a
relevant effect on reading performance anyway. Thus, one would think that little would be lost
by using a Gothic typeface average in legibility
rather than a highly legible Roman. Moreover, the survival of the fittest is thought to result in
the disappearance of the less legible typefaces without scientific intervention (Søgren, 1995). This would imply that a popular
design has become popular as a result of its functional qualities, but alas, it is impossible to
check this assumption.
