Baskerville is the "Typeface of Truth"

marcox's picture

That's the conclusion that award-winning filmmaker Errol Morris reached after a clever online experiment, reports Michael Bierut.

quadibloc's picture

Well, I was dismayed that they included Computer Modern in that test instead of Century Schoolbook or Century Expanded; this limits the usefulness of its results.

I would also have been very interested to learn where Times Roman ranks in such a test.

Before looking at the article in the link, though, this reminded me of another famous saying... but though I read it in a book, on the web, it seems that Tamazight or French are more often given the title of the language of free men.

Nick Shinn's picture

If the typeface can make a difference, how about leading?
Point size?

Perhaps Baskerville might not have seemed so truthful with a different typespec.


These kind of “priming” tests (to use Kahneman’s term) point in an interesting direction, but they are just the tip of the iceberg. As graphic designers, we work with the full range of typographically expressive means, and to suggest that merely choosing a particular font will have specific and predictable neurological effects is to banalize our profession.

David Vereschagin's picture

The informal study was only among six typefaces, hardly enough variety to claim that Baskerville has a lock on projecting authority and credulity. Looking at the other faces: : Computer Modern, Georgia, Helvetica, Comic Sans and Trebuchet, what the outcome says to me is that tradition is seen to equal truth, i.e. the oldest typeface looked the most trustworthy. Let’s see a contest among Baskerville, Caslon, Bodoni, Garamond and other pre-20th century faces and then see what happens. Let’s change the content to something non-scientific (in this case it was an asteriod hitting Earth) and see what happens. Let's change the colour of the text, change the size, the leading, the column width, etc. etc. and see what happens.


Nick Shinn's picture

This does suggest a new way for foundries to promote their wares.

Just test one’s new type against some old rubbish (MS core Truetype web fonts would do in a pinch) in a layout which favors it, then proclaim that science proves that one’s new design is the Best Ever.

ldavidson's picture

Yes, he could have expanded the test, and I agree that different sizes of fonts might affect the results; Trebuchet looked nasty on my screen on the size used, but much better enlarged a bit. The results are still interesting.

Seeing that the fonts were rendered by browsers, I wonder if they could find differences in results based on the browser used. It might have been better to use images. I assume browsers present them more uniformly.

Andreas Stötzner's picture

> Let’s see a contest among Baskerville, Caslon, Bodoni, Garamond and other pre-20th century faces and then see what happens.

This was exactly my thought.
Perhaps, if he had put Garamond or Times in place for Baskerville, we’d read now “Garamond is the … ”.

To let Baskerville compete against Comic Sans is like to let people choose wether the Smart or the BMW looks “like a car”.

Syndicate content Syndicate content