Safekipedia

Sound change

Discoverer · ages 10+Adapted from Wikipedia

In historical linguistics, a sound change is a change in the pronunciation of a language. A sound change can replace one speech sound, or one phonetic feature value, with a different one. This is called phonetic change. It can also be a broader change to the sounds that exist, such as two sounds merging or a new sound appearing. This broader kind is called phonological change. A sound change can remove a sound, or it can add a new one. Some sound changes only happen in certain sound environments and not in others.

The term "sound change" refers to diachronic changes, which happen over time in a language's sound system. "Alternation" is different. It refers to changes that happen synchronically, within the speech of one person, depending on the sounds nearby. Alternation does not change the language's underlying system. For example, the -s in the English plural is pronounced differently depending on the sound before it, as in bet[s] and bed[z]. This is alternation, not sound change. Sometimes "sound change" can also mean the historical start of an alternation, as with the /k/ sound in the Tuscan dialect. That label is not always precise, so it often needs to be explained as either a change in phonemes or a restructuring of the sound system.

Researchers usually assume that sound change is regular. This means it applies in a predictable way whenever its sound conditions are met, no matter what the words mean. Some words seem to break this rule because of dialects borrowing words, grammatical analogy, or other causes. Changes that affect only one or a few words, with no clear pattern, are called "sporadic."

The Neogrammarian linguists of the 19th century introduced the term sound law for regular changes, perhaps copying the laws of physics. The word "law" is still used for certain named sound changes, such as Grimm's law and Grassmann's law. Real sound laws often have exceptions. Still, expecting regularity is very useful as a heuristic, because it lets historical linguists define regular correspondence through the comparative method. Each sound change is also limited in space and time. It works in a certain area, often within particular dialects, and only for a certain period. For these reasons, some people have criticized the term "sound law" for suggesting a universal rule that sound change does not really follow.

Principles

Scholars who study sound change, especially those following the Neogrammarian model, use a few ideas as rules of thumb. Modern linguists see these ideas as guidelines rather than strict laws. Sound change has no memory: a new change cannot affect only one of two sounds that have already merged. Sound change ignores grammar: a change depends on sound patterns, such as sounds in unstressed syllables, and not on word types like adjectives. A sound change may recognise word boundaries even without prosodic clues. Changes may also be regularized in inflectional paradigms, where they become morphological rather than phonological.

Sound change is exceptionless means that if a change can happen somewhere, it affects every sound that fits. Apparent exceptions can come from analogy, other sound changes, or factors that have not been noticed. In recent decades, the theory of lexical diffusion has challenged this idea, arguing that change does not always affect every word. Still, once a change starts, it often spreads to the whole lexicon. For example, the Spanish fronting of Vulgar Latin [g] (voiced velar stop) before [i e ɛ] seems to have reached every possible word. The voicing of Latin [k] to [g] at the start of words happened in colaphus > golpe and cattus > gato, but not in canna > caña. Sound change is inevitable: all languages vary from place to place and over time, and neither writing nor media prevents that change.

Formal notation

See also: Phonological rule

A sound change is written as "A > B", which is read as "Sound A changes into sound B." A belongs to an older stage of a language, and B belongs to a more recent stage. For example, "Proto-Oceanic (POc.) *t is reflected as [f] in Rotuman (Rot.)." This kind of note shows only the start and end of a change, but there may have been steps in between. The *t sound first changed to [θ], like the start of English thin, and later became [f]. This can be written as t > θ > f.

When a change happens only in certain places, the surrounding sounds are written after a slash. "A > B / X__Y" means "A changes to B when it is preceded by X and followed by Y." For example, [b] inherited from Latin became [v] in Italian when it was between vowels. In Proto-Iranian, a voiceless stop changed into a fricative when it was followed by a continuant consonant, which is a resonant or a fricative. In the variety of Iranian that led to Old Persian, this change happened in all clusters. The symbol "#" marks a word boundary, so "__#" means word-final. In Greek, word-final stops were deleted, and P stands for any plosive.

In historical linguistics, several traditional terms describe kinds of sound change. Some changes happen only now and then, like accidents to one specific word. Others affect a whole sound system, which is described in phonological change. Assimilation is when one sound becomes more like another, as when the Latin prefix kom- became con- before an apical stop or [n]. Most assimilations involve neighboring sounds, and the earlier sound usually changes to match the later one. Dissimilation is the opposite, when one sound becomes less like another. Old Spanish Old Spanish omne "man" became hombre, and Grassmann's Law in Sanskrit and Greek is a regular example. Metathesis is when two sounds switch places, as in Old English thridda becoming third. Lenition is when a consonant becomes weaker, such as a stop consonant becoming an affricate or fricative. Its opposite, fortition, makes a consonant stronger, such as an approximant becoming an affricate or fricative. When vowels weaken, the change is called reduction, as when unstressed vowels often become a schwa.

Other changes include tonogenesis, where syllables come to have distinctive pitch contours. Sandhi describes changes that happen only at word boundaries, and it can be specific to a morpheme. Haplology is when a syllable is lost because a neighboring syllable is similar. Elision, aphaeresis, syncope, and apocope are all losses of sounds, affecting unstressed, initial, middle, and final sounds. Prothesis adds a sound at the beginning of a word, and nasalization makes vowels nasal. If the nasal consonant is lost but the vowel stays nasal, the nasalization becomes phonemic, meaning it is distinctive, as in French words ending in "-in". Over time, the sounds in assimilations next to each other were usually regular changes as well, as explained in diachronically speaking. In liaison, the [n] is still pronounced in some cases.

Examples of specific sound changes in various languages

Keep exploring

This article is a child-friendly adaptation of the Wikipedia article on Sound change, available under CC BY-SA 4.0.