by Konstantinos Petridis
A suggestion
Two years ago, at Typo Thessaloniki, I had the good fortune to meet Erik Spiekermann, one of the most important living type designers. The purpose of that meeting was to draw on his experience to help us with the design of Greek typefaces we were then working on, aiming to offer the Greek market fonts under the brand name “Locomotive Fonts”—a project that was later abandoned due to lack of time.
We had the impression that the appearance of a Greek text (textus), regardless of the typeface, tends to look more uneven, more restless, and therefore less legible than the corresponding Latin text, but we couldn’t identify what caused this.
From that discussion arose a shared perspective and a joint project that I would call “Greek Phonemic Writing.”
After two years of studying the evolution of the Greek alphabet—from its origins to the present day—and with the help of linguists, Spiekermann and I arrived at a proposal that was finalized in the spring of 2006 in Berlin.
The study showed that the problem lies primarily within the typographic characters themselves—especially those inherited from the Renaissance tradition of the first versions of Greek lowercase type. In those early designs, the main goal was not legibility, but rather resemblance to the handwritten forms of the calligraphers of the time, such as the great calligrapher Angelos Vergikios.
This happened because, in its early stages, typography aimed to be a more affordable version of the single handwritten manuscript of the Middle Ages. Type cases containing more than 1,000 different sorts were not an exception. The lowercase “alpha” alone had twelve variations (plain alpha, alpha with acute, alpha with grave, alpha with rough breathing, alpha with circumflex, alpha with smooth breathing and acute, etc.). The ligatures were so numerous that special explanatory charts were needed for the typesetters.
Κάτω: Πίνακας εnεξήγnσnς των συμπλεγμάτων και τnς θέσnς τους στnν τυπογραφική κάσα
The “lowercase” letters, with the change of the writing medium (the pen), had already prevailed by the 9th century AD, because they were written more quickly and read more easily than “uppercase” letters.
Adrian Frutiger, seeking to scientifically explain the reason for the dominance of lowercase over uppercase writing, analyzed the movements of the pen in forming a letter. Thus, according to Frutiger, the degrees of difficulty are analyzed as follows:
The uppercase “A” scores 15 degrees of difficulty, while the lowercase “a” only 5. If we subject the Greek lowercase typographic characters to a Frutiger-style analysis, we will find that their degree of difficulty far exceeds that of the corresponding Latin letters.
It is therefore clear that Greek lowercase letters are both harder to write and harder to read than their Latin counterparts. What is so special about the Latin ones? The Latin characters have been influenced by all the modern artistic movements and have reached a high degree of simplification, economy, and visual clarity—something that has not happened with the Greek characters, which still rely, more or less, on the original designs of the Renaissance.
One symbol, four elements. Economy and simplification without any compromise in clarity or recognizability.
The difficulty in Greek characters is found in letters that have always had a high degree of difficulty in written form, such as “ζ”, “ξ”, “φ”, “Ψ”, and the left-leaning “λ”.
A comparison of the visual appearance of texts (textus) clearly favors the Latin ones.
Those are the issues regarding the degree of difficulty in writing and reading Greek characters. As for their correspondence with the phonetic system of Modern Greek, things are even more problematic.
We consider it our duty to use the same system of symbols that our ancient ancestors used, even though pronunciation has undergone dramatic changes from then until today. As a result, we are forced to invent improbable ways to explain the pronunciation of a single symbol — for example, of “upsilon” (the pronunciation of the letters will be indicated with Latin characters in brackets).
We say that
“υ” by itself is pronounced [i],
“υ” together with “ι” is pronounced [i],
“υ” together with “ο” is pronounced [u],
“υ” following “ε” or “α” and before a voiceless fricative consonant such as “φ”, “χ”, or “θ” is pronounced [f],
“υ” following “ε” or “α” and before a voiced fricative consonant such as “β”, “γ”, or “δ” is pronounced [v],
and when the diphthong is stressed, we even have to place the accent mark on the consonant!
For example, “εύκολος” (the accent on “υ,” which has the phonetic value of a consonant [f]).
This truly is a curse. It could, however, turn into a blessing if we could adopt the spirit rather than the letter of our ancient ancestors. Let me explain what I mean. Among other things, our ancestors were pioneers in this field as well.
They invented independent visual symbols for the vowels, in contrast to the Phoenician script they had adopted, which was a consonantal alphabet suited to Semitic languages (where vowels were not written because they were predictable).
We know that the vowels of Ancient Greek were distinguished as short and long.
However, since we know that every writing system is imperfect and cannot capture spoken language with 100% accuracy, we see that there was initially no clear distinction between the short “ε” and the long “ε,” that is, “η.” Thus, in one of the earliest inscriptions written with the “Phoenician letters” (as Herodotus calls them), on Nestor’s Cup, the word appears as ΠΟΤΕΡΙΟΝ instead of ΠΟΤΗΡΙΟΝ, and later as ΗΕΧΣΕΚΟΝΤΑ instead of ΕΞΗΚΟΝΤΑ.
In the Decree of the Athenian Demos, we also see the letter “Η” used in place of the rough breathing [h], and the letter “Ξ” correctly composed of two phonemes, “Χ” and “Σ,” represented by two different symbols, where “Χ” has the phonetic value [kh]. The letter “Ε” carries the phonetic value of both the short “ε” and the long “η.”
Later, when the rough breathing [h] ceased to be pronounced, “Η” stopped being written in that position and was reassigned to represent the long “ε,” giving rise to the later spelling ΕΞΗΚΟΝΤΑ.
The same thing happened with the letter “san” (represented by the symbol Μ), a symbol for the aspirated “σ” [sh], which disappeared from inscriptions as early as the archaic period because it was no longer pronounced. In some cities, it did not even appear as a symbol in their alphabet because its pronunciation had already disappeared before the adoption of the “Phoenician” alphabet.
Whatever our ancestors no longer pronounced, they stopped writing—or they reused it elsewhere to clarify pronunciation details that had previously been insufficiently indicated.
As simple as that.
Why shouldn’t we do the same?
Why shouldn’t we write exactly what we pronounce?
You might say, how would we be able to distinguish the etymological difference between “εφορία” and “ευφορία”, which under the new system would be written the same way? Etymology is the linguists’ sport, and we don’t all need to become linguists in this country.
We would simply learn that there are two kinds of “eforía”, with different meanings — just as we already know that there are two or more meanings for words such as “kanonáki” (a small cannon or a traditional musical instrument), “týpos” (a mathematical formula, a type of person, or the daily press), “télos” (a fee or the end of innocence), “róda” (the wheel and the roses), “pláti” (the back and the breadths), “báza” (the trick in cards and the construction rubble), “váta” (the plant and the cotton), “péta” (the petals and the imperative of I fly), “métrα” (taking measures, a unit of length, and the imperative of I measure), inferring the difference from context — just as we do in spoken language.
Of course, over time there have been some attempts to simplify spelling, such as “treno” (train), “avgo” (egg), “etairia” (company), as well as the elimination of double consonants in foreign words such as “kaseta” (cassette) and “afisa” (poster). But these efforts were fragmentary and limited, as they met with strong resistance.
After all these reflections, Spiekermann and I arrived at a proposal that would coexist alongside the other Greek alphabets, such as:
- Polytonic (historical and simplified)
- Monotonic (official and unofficial versions)
- Atonic (my sons have been using it since first grade)
and ~ - Greeklish (various personalized versions)
We therefore propose a Greek phonetic alphabet with 20 letters, as follows:
α, β, γ, δ, ε, z, θ, ι, κ, Λ, μ, ν, ο, π, ρ, s, τ, υ, φ, χ
where we have the following substitutions:
instead of
ου → υ (the “upsilon” regains its ancient phonetic value [ou])
ξ → κs, which will consist of two symbols [k] + [s] and will appear on the keyboard in place of the old “ξ” (for greater typing economy)
ψ → πs, which will consist of two symbols [p] + [s] and will appear on the keyboard in place of the old “ψ”
η, ι, οι, ει, υι → ι
ω, ο → ο
ε, αι → ε
λ → Λ (for ν and Λ, symbol economy) and elimination of the upward stroke, which forces the hand to return leftward before continuing with the first part of the letter.
We part ways with the difficult letter “ξ” and use the capital “z” for “ζ,” as is done in Latin and already in several Greek typefaces, to reduce its complexity. A single symbol for sigma, the “s,” replaces both “σ” and “ς,” and we also remove the descenders from “χ.”
We also propose the elimination of double consonants, since they have long ceased to be pronounced as doubles.
The problem remains, however, with the ambiguity of “ντ,” “γκ” (“γγ”), and “μπ,” which in the current system are unclear as to whether they should be pronounced [b], [d], [g] or [mb], [nd], [ng]. This ambiguity will continue to exist, because a significant linguistic change is taking place in our time, right before our eyes and ears. There is a strong tendency toward the disappearance of prenasalization in [b], [d], and [g]—that is, the loss of the [mb], [nd], [ng] pronunciation and its replacement with [b], [d], [g], mainly in southern Greece but gradually in the north as well.
Where can this be observed? Here’s a visual example:
Two different people ordered these two signs. One pronounces the word “Λαγκαδάς” as Langadas, and the other as Lagadas. There is also a third, bureaucratic version, Lagkadas, where the logic is to transliterate Greek into Latin letter by letter. The result: confusion. These signs currently stand one behind the other just before Lagadas for those approaching from Kavala. Imagine the poor tourist, for whom these signs are made, holding a map that might even contain a fourth, previously unknown version of Lagadas.
Anyone who keeps a distance from attempts to ideologize Greek writing can try out the new typeface. It is Spiekermann’s Meta typeface, available in two weights (normal and bold), both upright and italic. It will be available on the FontShop website from October 2006, at https://www.fontshop.com/, for anyone to use freely.
It will also be available for free from LineaDesign
Contact us
In closing, I include a text written in the traditional way, and the same text written according to the new logic, albeit using an existing typeface.
