An alphabet is a small set of letters--basic written symbols--each of which roughly represents or represented historically a phoneme of a spoken language. This is distinguished from other writing systems such as ideograms, in which symbols represent complete ideas, and syllabaries, in which each symbol represents a syllable.
Each language may establish certain general rules that govern the association between letters and phonemes, but, depending on the language, these rules may or may not be consistently followed. In a perfectly phonological alphabet, the phonemes and letters would correspond perfectly in two directions: a writer could predict the spelling of a word given its pronunciation, and a speaker could predict the pronunciation of a word given its spelling. However, languages often evolve independently of their writing systems, and writing systems have been borrowed for languages they were not designed for, so the degree to which letters of an alphabet correspond to phonemes of a language varies greatly from one language to another and even within a single language.
Languages may fail to achieve a one-to-one correspondence between letters and sounds in any of several ways:
- A language may represent a given phoneme with a combination of letters rather than just a single letter.
- A language may represent the same phoneme with two different letters or combinations of letters.
- A language may spell some words with unpronounced letters that exist for historical or other reasons.
- Pronunciation of individual words may change according to the presence of surrounding words in a sentence.
- Different dialects of a language may pronounce different phonemes for the same word.
National languages generally elect to address the problem of dialects by simply associating the alphabet with the national standard. However, with international languages with wide variations in its dialects, such as English, it would be impossible to represent the language in all its variations with a single phonetic alphabet.
Some national languages like Finnish and Spanish have a very regular spelling system with close to a one-to-one correspondence between letters and phonemes. In standard Spanish, it is possible to predict the pronunciation of a word from its spelling, but not vice versa; this is because certain phonemes can be represented in more than one way, but a given letter is consistently represented. French, with its silent letters and its heavy use of nasal vowels and elision, may seem to lack much correspondance between spelling and pronunciation, but its rules on pronunciation are actually consistent and predictable with a fair degree of accuracy. At the other extreme, however, are languages such as English, where the spelling of many words simply has to be memorized as they do not correspond to sounds in a consistent way (though they may have at some earlier time in the language's evolution). However, even English has general rules that predict pronunciation from spelling, and these rules are successful a majority of the time.
The first alphabet was probably developed by the Canaanites around 1700-1500 BC (see early Semitic alphabet), and nearly all subsequent alphabets are derived from it or inspired by it, directly or indirectly. Of special note among its descendants is the Greek alphabet, which was the first to have separate symbols for vowels (Semitic didn't need them). Most subsequent alphabets with vowels are derived from the early Greek alphabets. The most popular alphabet in use today is a modern 26-letter version of the Roman alphabet, used by the English language and most European languages. In modern linguistic usage, the term latin alphabet is usually used to refer to the modern derivations from the alphabet used by the Romans (i.e. the Roman alphabet).
An alphabet also serves to establish an order among letters that can be used for sorting entries in lists, called collating. Languages that use the roman alphabet have varying collating rules: In German and English, umlaut characters (, , ) are treated just like their non-umlauted versions, which makes the alphabetic order ARG, RGERLICH, ARM. In Swedish, W is a foreign letter which is treated just like V, but the alphabet has three extra letters at its end (..., X, Y, Z, , , ), giving the order URBAN, WALTER, VILGOT, KE, RLIG, STEN. This order is different in Danish and Norwegian (..., X, Y, Z, AE, , OE). Some languages have more complex rules: for example, Spanish treats "ch" and "ll" as single letters, giving an ordering of CINCO, CREDO, CHISPA and LOMO, LUZ, LLAMA. In Icelandic, Þ is added, and D is followed by Ð. Both letters were also used by Anglo-Saxon scribes who also used the Runic letter Wynn to represent /w/. Þ (called thorn; lowercase þ) is also a Runic letter, some scholars derive it from Latin D. Ð (called eth; lowercase ð) is the letter D with an added stroke.
In recent years the Unicode initiative has attempted to collate most of the world's known alphabets into a single character encoding. As well as its primary purpose of standardising computer processing of non-Roman scripts, the Unicode project has provided a focus for script-related scholarship.
The sounds of speech of all languages of the world can be written by a rather small universal phonetic alphabet. A standard for this is the International Phonetic Alphabet
- Morse code
- NATO Phonetic
- Old Turkic
- Old Italic