Let Us Build a Tower: Unicode and The Babelization of ASCII

You can find explanations of Unicode all over the web, this explanation is offered for non-technical people in the field of biblical studies. The purpose is to help you understand what you need on your computer to be able to enter text in German, French, Hebrew, Greek, or other languages with special characters. Specifically, it will help you understand why no ordinary font will do.

Let Us Build a Tower

Back when I got my TI-99/4A (4k RAM–twice as much as my friend’s Timex Sinclair!) I learned about ASCII, which is the system of numeric codes used to identify the characters available via the keyboard (commands, letters, numbers, punctuation, non-printing characters, etc.). Since computers understand numbers (0 and 1) and not letters, keyboard characters need to be translated into numeric codes in order for the computer to understand them. Originally, the ASCII system was set up to handle seven bits worth of characters. That’s 128 characters. Since 33 of those characters were command, or non-printing characters, that left just 95 for numbers, punctuation, and both small and capital letters.

That’s fine for English, but what if you want to type in a language other than English? In a lot of languages, 128 won’t be enough. But let’s go further. Suppose you want to have all the languages of the world encoded in the same system? You’d need a lot more than 7 bits and 128 characters to be able to do that.

In fact, this was the original vision of Unicode (a descendant of ASCII): “Let us build a consistent system of encoding all the characters of all the world from the beginning of time.” That’s a pretty ambitious goal, but they’re getting closer. The current version of Unicode (5.2 at this time of writing) accommodates 107,000 character encodings!

What does this have to do with you? Think back to the old Bibleworks or Logos or Gramcord fonts for biblical Greek and Hebrew. It used to be that when you typed something into your word processor using one of those fonts for Hebrew, and then changed the font to something like Arial, the text would change from Hebrew characters into some Latin based gobbledygook. That shouldn’t happen. In an ideal world, you should be able to change the font and your Hebrew text should stay in Hebrew. That way you can send your file to someone who doesn’t have the same fonts installed and they would still be able to read your document. You can’t do that in ASCII because there’s only room for 128 characters. That’s why Unicode eventually evolved out of the original ASCII system.

Maybe this will help things become even more clear. In a custom encoded font (the old ASCII system), when you hit the letter “a” on your computer, the computer recognized that letter according to the numeric code for “a,” but out on the screen popped the letter alef. That happens because the font distorts the “a” to make it look like an alef (if I may speak metaphorically). With a Unicode font, on the other hand, when you type the letter “a”, the computer recognizes the numeric code for alef and out on the screen pops the letter alef. That’s really good because now what you have on your screen is really truly an alef–a character that not only looks like alef but is actually encoded as alef, not just an “a” masked as an alef.

Going back to the example I just gave where I type an “a” and my computer reads that as the numeric code for “alef,” you may wonder how my computer knows that I mean alef and not “a.” With Unicode, I may not need a special font to type in German or French or even Hebrew, but I do need special software that will remap my keyboard so that I can type German or French or Hebrew characters instead of English characters. Theoretically then, if I want to type in Hebrew, all I need is software to remap my keyboard according to the Hebrew block of the Unicode character map and I’m all set. I can type in Hebrew in any Unicode font and it will show up in beautiful Arial, Times New Roman, or California Hebrew–theoretically.

Theory and Reality: Keyboards and Fonts

I can imagine it would be pretty difficult to design a font that would be graphically pleasing and consistent from one character to the next just within my own language with all its punctuation and numbers and letters. Then to imagine that I need to do that for every language on the face of the earth–living or dead!? Remember there’s room for 107,000 characters in the Unicode system. And some of those writing systems do some pretty funky things–like go right to left, or top to bottom, or pile up on top of each other (like accents or Hebrew vowels). So it’s no wonder that fonts often get designed to accommodate specific language groups (Latin or Cyrillic) and not for every writing system on the face of the planet. One font, called Code2000 has an exceptionally large area of the Unicode map covered. That is, less than 6,500 of the 107,000 possible.

I hope that makes it clear why there are two things to think about when you want to input Hebrew, Greek, transliteration codes, or modern foreign languages into your dissertation document. You need key mapping software that will remap your keyboard to access the proper block of the Unicode character map (the special Latin characters for German or French, the Ugaritic section, the IPA section, etc.) and you need a font that will properly display the characters you want to type.

The Dissertation Template

In terms of German and French and other modern languages. These are easily accommodated by Times New Roman (default for the template) and other standard fonts. The keyboards for these languages are also available with standard versions of Microsoft Windows. These languages should not create any problems.

Hebrew and Greek are also easy to implement. I recommend downloading the Tyndale Unicode Font Kit. This kit includes both keyboards and fonts for Greek and Hebrew. I recommend using the keyboards, but not the fonts. The template uses SBL Hebrew and Gentium, now I recommend using SBL BibLit which now works with just about any text you may want to use for biblical studies.

Those challenges are easily solved. More difficult is the problem of transliterating ancient scripts. I’ll cover that in a separate post.