Most tibetologists use a system devised by Turrell Wylie some forty years ago (A standard system of Tibetan transcription, Turrel Wylie, in: Harvard journal of Asiatic studies 22 (1959) p. 261-267). The advantage of Wylie’s transliteration is that letter combinations instead of diacritics are used for distinction. A weak point, however, lies in the fact that the twenty-third ('a) and the thirtieth letter of the alphabet (a) are not expressed by what they actually are in the logic of Tibetan writing, i.e. consonants. Therefore, Wylie’s system has been slightly modified to render the transliteration table for the thirty Tibetan letters as follows:
The use of v for the so-called vachung is a convention widespread in China. The thirtieth letter appears in words like xamdo “Amdo”. The apostrophe is used in cases like n‘yagrota “fig tree” where n‘y is not the eighth letter of the alphabet but a combination of n plus y. In a similar way, the apostrophe is used for marking subscript h in older Tibetan orthography (rdzogs s‘ho) and initials involving g and y in such words like g‘yag “yak” versus gyang “wall”. The apostrophe is eliminated in the indexes. Please also note that Tibetan words which consist of more than one syllable are written as words without using devices like the hyphen for separating the syllables. This is absolutely essential in a catalogue for assuring an adequate indexation of the keywords and avoidance of trash entries (for a detailed argumentation please see here: On letters, words, and syllables, transliteration and romanization of the Tibetan script, by Michael Balk.
Here is a short description of how wording (or, word spelling) is effected: Tibetan syllables can be classified into stems (or elementary syllables) and enclitics. Stems can be found at the beginning of a sentence while enclitics cannot be used initially. They are usually called particles and will always follow another syllable. A rough distinction is one between nominal and other particles. Nominal particles are used for the formation of nouns while the other particles are relevant for expressing syntactical relations of various kinds. With this, we arrive at three categories of syllables indicated by the letters A, B, and C:
A stems (chos, rgyal, bstan, etc.) B nominal particles (pa, mo, kha, can, etc.) C other particles (kyi, du, cing, dag, ste, so, etc.)
A stem is only written as a single word if it represents a one-syllable lexical entity or a final verb. In other cases up to three stems may be written as a word if they form a habitual syllable compound. This yields these types:
A chos “Religion”
AB rgyalpo “king” (rgyal-po) AAB mustegspa “heretic” (mu-stegs-pa) AAAB bcomldanvdasma “Bhagavati” (bcom-ldan-vdas-ma, name of a sutra)
Some nouns consist of stem + nominal particle + another stem. You would have it in:
ABA rinpoche “incarnated lama” (rin-po-che)
Another possibility is a junction of more than one nominal particle as in:
ABB rtsompapo “author” (rtsom-pa-po)
AABB lasdangpoba “beginner” (las-dang-po-ba)
For particles other than nominal particles, a separate spelling is used. When looking for particular terms or concepts in an alphabetical index one would normally not entertain entries increased by inflectional distinctions such as cases or other purely syntactical information. There are only two cases where it is natural to connect one of the other particles, namely case particles, with a preceding syllable as an exception to the rule. One is a number of adverbs. The other is if case particles are used within personal names:
AC rabtu “very” and the like (rab-tu)
AC Choskyi “Choekyi” (chos-kyi)
Word spelling of the rabtu type is restricted to two-syllable adverbs. We write rabtu vbyungba “to enter asketic life” but lag tu vjugpa “to hand over” and pharol tu phyinpa “supremacy”. The spelling Choskyi implies that the term is part of a name, while “sphere of religion” is chos kyi dbyings.