Anyone who lives in Israel or has visited knows that many, perhaps most, road signs are written in both Hebrew and English (and often in Arabic as well). The English signs - meaning those written in Latin characters - are not translations of the Hebrew, but transliterations. Those who don't read Hebrew can't easily figure out how to pronounce place names simply by reading the Latin transliterations. It's not easy to guess the proper pronunciation of any foreign language. But signs of this sort are not intended for foreign tourists alone. Hundreds of thousands of Israelis look at signposts and see the Latin characters first, even if many of them know some Hebrew. Some of them are newcomers, others veteran immigrants who came as adults and haven't perfected their Hebrew. Others are foreign workers or members of the diplomatic corps. For some, Latin transliterations offer a "moment of Hebrew" which helps non-Hebrew readers make some connection between the Hebrew and English alphabet. This is a strong reason why the Latin characters of Hebrew words should be accurately written and clearly represent the Hebrew original. For example, Hebrew has two letters for the "K" sound. Certain words need to be transliterated with a "K" and others "Q," - for instance, "Ksalon," but not "Qoranit." Latin characters that do not identify an individual Hebrew phoneme - that is, a unit of a sound in a language - with a unique transliterated letter do not help non-Hebrew readers, and are actually confusing to those who have some command of Hebrew. THE ACADEMY for the Hebrew Language, of which I am a member, is now addressing the problem of Latin transliterations of Hebrew signpost names. We've been under a great deal of pressure to introduce a standard for converting Hebrew words into English. This pressure comes from those involved with tourism and transportation. I believe the way to proceed is to convert each Hebrew phoneme into a separate and different Latin letter. Hebrew speakers know the difference between ayin and aleph in writing, even though many people pronounce them identically. For a tourist the difference may be insignificant, but Israelis know the distinction is essential. Similarly, in preparing a conversion of a Hebrew phoneme into separate Latin letters, the academy should separate between tav and tet. They obviously should be converted into a distinct Latin letter. Tet should be written with a dot under it, while tav should be a simple "t." The main feature that the Standards Institution of Israel and its Conversion of Scripts office demands is that it be possible to automatically reconstruct a converted text into the original script. That is the criterion we should uphold for our signposts. Another principle for transliteration into Latin is that each Hebrew phoneme be converted into only one letter, not a combination of letters. The solution - as with most European languages transliterated into English - lies in employing a diacritical or accent mark such as small line under a letter, or small "v" above a letter. There is a third point. "G" and "c" in English and many other languages have each two pronunciations. Still, both are represented by the same letter. Hebrew also has letters that have each two pronunciations. They are "bet", "pay", and "kaph" which sometimes have the phonetic value of "v," "f" and the Spanish "x," respectively. Hebrew speakers know how to pronounce them properly. They are essential to Hebrew writing. But for signposts I suggest that a small line be added under the letter to show the intended pronunciation. In the final analysis, no nation should distort the elements of its language just for the sake of making life easier for tourists. Greater issues are at stake. We have every reason to expect that no tourist will get lost if the formulation I advocate for transliterations is adopted. The writer is a full member of the Academy for the Hebrew Language, a professor emeritus at the Hebrew University of Jerusalem and visiting professor at the Faculty of Computer Science at the Technion.