In this work I have attempted to apply the techniques used in constructing biological field guides to the problem of identifying typefaces. I believe that this approach ma kes matters simpler for novices, because at each step it directs their attention to th e most salient characteristics of the item being identified.
It is important to understand the relationship between field guides and taxonomy. In taxonomy, an individual is a point in a multidimensional space. For example, if we conservatively assume their are ten distinguishing features per letter, and if we restrict ourselves to alphabetic characters, a typeface such as "Caledonia" can be viewed as a point in a 520-dimensional space. Higher-level taxa like "Modern Serif" are volumes in that space. The role of taxonomy is to discover relationships and affinities among the individuals and taxa within that entire multi-dimensional space. Attempting to keep all the dimensions in mind at once is very difficult, however. The solution used by field guides is to focus the attention on just one or two dimensions at a time, and to use the dimensions that seem easiest to discriminate on, whether they are viewed as being taxonomically the most significant or not. Thus the author of a field guide is free to use categories such as "pink flowers", while the taxonomic botanist is apt to concentrate on whether the pistils are adnate to the perianth. Of course, a good field guide will use important taxonomic discriminators whenever possible, but when a choice must be made, the field guide will opt for the characteristic that is easiest for the beginner to use. I feel that the failure to exploit this distinction between taxonomy and field guides has hampered most of the typographic identification manuals in the past. Such manuals usually just present a few high-level categories and then assume that users will wade through many pages of specimens to find the face they are looking for. The Rookledge Typefinder adds an abundance of earmarks, but they are not organized in a way that would let the novice step through them systematically. Bauermeister's Manual of Comparative Typography takes a contrastive approach, but tries to make do with just 7 dimensions - with the result that it makes unreasonable demands on the novice.
In The Field Guide to the Faces I have attempted to avoid features that are difficult for novices to judge or which, in these days of Multiple Master fonts, are frequently varied within font families. Starting with the largest groupings and working down through smaller and smaller taxa, I have at each point tried to discriminate based on the most obvious features of the group at hand, without preconceived notions of type history or aesthetics. When no sufficiently clear distinguishing feature presented itself, I have not hesitated to lump two or more faces in a single terminal class.
It should be pointed out that the approach I have taken suffers from a problem that affects all such contrastive methods: it is not robust in the face of incomplete data. Just as the user of a botanical field guide is often frustrated when he or she sees a very distinctive leaf pattern on a plant that is not in bloom, and the field guide asks only about the floral structure, so too you will be out of luck if your specimen has no "W" and yet my guide uses that letter to make a discrimination.
Ultimately, of course, what one would like to do is to use numeric methods and cladistics to assess the true affinities and relationships among typefaces. One can imagine the day when the family tree of typefaces rests on a scientific basis, and perhaps a whole new scientific nomenclature, analogous to Linnaean nomenclature in biology, replaces our current ad hoc names. I would like to think that the present volume is a first small step in that direction.
I have a tool which can identify typefaces from scanned specimens.