In linguistics, and particularly phonology, stress or accent is the relative emphasis or prominence given to a certain syllable in a word or to a certain word in a phrase or sentence. That emphasis is typically caused by such properties as increased loudness and vowel length, full articulation of the vowel, and changes in tone. The terms stress and accent are often used synonymously in that context but are sometimes distinguished. For example, when emphasis is produced through pitch alone, it is called pitch accent, and when produced through length alone, it is called quantitative accent. When caused by a combination of various intensified properties, it is called stress accent or dynamic accent; English uses what is called variable stress accent.
Since stress can be realised through a wide range of phonetic properties, such as loudness, vowel length, and pitch (which are also used for other linguistic functions), it is difficult to define stress solely phonetically.
The stress placed on syllables within words is called word stress or lexical stress. Some languages have fixed stress, meaning that the stress on virtually any multisyllable word falls on a particular syllable, such as the penultimate (e.g. Polish) or the first (e.g. Finnish). Other languages, like English and Russian, have variable stress, where the position of stress in a word is not predictable in that way. Sometimes more than one level of stress, such as primary stress and secondary stress, may be identified.
The stress placed on words within sentences is called sentence stress or prosodic stress. That is one of the three components of prosody, along with rhythm and intonation. It includes phrasal stress (the default emphasis of certain words within phrases or clauses), and contrastive stress (used to highlight an item, a word or part of a word, that is given particular focus).
There are various ways in which stress manifests itself in the speech stream, and these depend to some extent on which language is being spoken. Stressed syllables are often louder than non-stressed syllables, and may have a higher or lower pitch. They may also sometimes be pronounced longer. There are sometimes differences in place or manner of articulation - in particular, vowels in unstressed syllables may have a more central (or "neutral") articulation, while those in stressed syllables have a more peripheral articulation. Stress may be realized to varying degrees on different words in a sentence; sometimes the difference between the acoustic signals of stressed and unstressed syllables are minimal.
These particular distinguishing features of stress, or types of prominence in which particular features are dominant, are sometimes referred to as particular types of accent - dynamic accent in the case of loudness, pitch accent in the case of pitch (although that term usually has more specialized meanings), quantitative accent in the case of length, and qualitative accent in the case of differences in articulation. These can be compared to the various types of accent in music theory. In some contexts, the term stress or stress accent is used to mean specifically dynamic accent (or as an antonym to pitch accent in its various meanings).
A prominent syllable or word is said to be accented or tonic; the latter term does not imply that it carries phonemic tone. Other syllables or words are said to be unaccented or atonic. Syllables are frequently said to be in pretonic or post-tonic position; certain phonological rules apply specifically to such positions. For instance, in American English, /t/ and /d/ are flapped in post-tonic position.
In Mandarin Chinese, which is a tonal language, stressed syllables have been found to have tones realized with a relatively large swing in fundamental frequency, while unstressed syllables typically have smaller swings. (See also Stress in Standard Chinese.)
Stressed syllables are often perceived as being more forceful than non-stressed syllables.
Lexical stress, or word stress, is the stress placed on a given syllable in a word. The position of lexical stress in a word may depend on certain general rules applicable in the language or dialect in question, but in other languages, it must be learned for each word, as it is largely unpredictable. In some cases, classes of words in a language differ in their stress properties; for example, loanwords into a language with fixed stress may preserve stress placement from the source language, or the special pattern for Turkish placenames.
In some languages, the placement of stress can be determined by rules. It is thus not a phonemic property of the word, because it can always be predicted by applying the rules.
Languages in which the position of the stress can usually be predicted by a simple rule are said to have fixed stress. For example, in Czech, Finnish, Icelandic and Hungarian, the stress almost always comes on the first syllable of a word. In Armenian the stress is on the last syllable of a word. In Quechua, Esperanto, and Polish, the stress is almost always on the penult (second-last syllable). In Macedonian, it is on the antepenult (third-last syllable).
Other languages have stress placed on different syllables but in a predictable way, as in Classical Arabic and Latin, where stress is conditioned by the structure of particular syllables. They are said to have a regular stress rule.
Statements about the position of stress are sometimes affected by the fact that when a word is spoken in isolation, prosodic factors (see below) come into play, which do not apply when the word is spoken normally within a sentence. French words are sometimes said to be stressed on the final syllable, but that can be attributed to the prosodic stress that is placed on the last syllable (unless it is a schwa, when stress is placed on the second-last syllable) of any string of words in that language. Thus, it is on the last syllable of a word analyzed in isolation. The situation is similar in Standard Chinese. French (some authors add Chinese) can be considered to have no real lexical stress.
Languages in which the position of stress in a word is not fully predictable are said to have phonemic stress. For example, English, Russian, Italian, Portuguese and Spanish. Stress is usually truly lexical and must be memorized as part of the pronunciation of an individual word. In some languages, such as Spanish, Portuguese, Lakota and, to some extent, Italian, stress is even represented in writing using diacritical marks, for example in the Spanish words célebre and celebré. Sometimes, stress is fixed for all forms of a particular word, or it can fall on different syllables in different inflections of the same word.
In such languages with phonemic stress, the position of stress can serve to distinguish otherwise identical words. For example, the English words insight and incite are distinguished in pronunciation only by the fact that the stress falls on the first syllable in the former and on the second syllable in the latter. Examples from other languages include German umschreiben (['ma?bn] "to rewrite" vs. [m'a?bn?] "to paraphrase"); and Italian ancora (['a?kora] "anchor" vs. [a?'ko:ra] "more, still, yet").
In many languages with lexical stress, it is connected with alternations in vowels and/or consonants, which means that vowel quality differs by whether vowels are stressed or unstressed. There may also be limitations on certain phonemes in the language in which stress determines whether they are allowed to occur in a particular syllable or not. That is the case with most examples in English and occurs systematically in Russian, such as (['zam?k], "castle") vs. ([z?'mok], "lock"); and in Portuguese, such as the triplet sábia (['sabj?], "wise woman"), sabia ([s?'bi?], "knew"), sabiá ([s?'bja], "thrush").
Dialects of the same language may have different stress placement. For instance, the English word laboratory is stressed on the second syllable in British English (labóratory often pronounced "labóratry", the second o being silent), but the first syllable in American English, with a secondary stress on the "tor' syllable (láboratory often pronounced "lábratory"). The Spanish word video is stressed on the first syllable in Spain (vídeo) but on the second syllable in the Americas (video). The Portuguese words for Madagascar and the continent Oceania are stressed on the third syllable in European Portuguese (Madagáscar and Oceânia), but on the fourth syllable in Brazilian Portuguese (Madagascar and Oceania).
With very few exceptions, English compound words are stressed on their first component. And even such exceptions, for example mankínd, are instead often stressed on the first component by some people or in some kinds of English. Sometimes the same components as those of a compound word are used in a descriptive phrase with a different meaning and with stress on both words, but that descriptive phrase is then not usually considered a compound: bláck bírd (any bird that is black) and bláckbird (a specific bird species) and páper bág (a bag made of paper) and páper bag (very rarely used to mean a bag for carrying newspapers but is often also used to mean a bag made of paper).
Some languages are described as having both primary stress and secondary stress. A syllable with secondary stress is stressed relative to unstressed syllables but not as strongly as a syllable with primary stress. As with primary stress, the position of secondary stress may be more or less predictable depending on language. In English, it is not fully predictable, but the different secondary stress of the words organization and accumulation (on the first and second syllable, respectively) is predictable due to the same stress of the verbs órganize and accúmulate. In some analyses, for example the one found in Chomsky and Halle's The Sound Pattern of English, English has been described as having four levels of stress: primary, secondary, tertiary, and quaternary, but the treatments often disagree with one another.
Peter Ladefoged and other phoneticians have noted that it is possible to describe English with only one degree of stress, as long as prosody is recognized and unstressed syllables are phonemically distinguished for vowel reduction. They find that the multiple levels posited for English, whether primary-secondary or primary-secondary-tertiary, are not phonetic stress (let alone phonemic), and that the supposed secondary/tertiary stress is not characterized by the increase in respiratory activity associated with primary/secondary stress in English and other languages. (For further detail see Stress and vowel reduction in English.)
Prosodic stress, or sentence stress, refers to stress patterns that apply at a higher level than the individual word - namely within a prosodic unit. It may involve a certain natural stress pattern characteristic of a given language, but may also involve the placing of emphasis on particular words because of their relative importance (contrastive stress).
An example of a natural prosodic stress pattern is that described for French above; stress is placed on the final syllable of a string of words (or if that is a schwa, the next-to-final syllable). A similar pattern is found in English (see § Levels of stress above): the traditional distinction between (lexical) primary and secondary stress is replaced partly by a prosodic rule stating that the final stressed syllable in a phrase is given additional stress. (A word spoken alone becomes such a phrase, hence such prosodic stress may appear to be lexical if the pronunciation of words is analyzed in a standalone context rather than within phrases.)
Another type of prosodic stress pattern is quantity sensitivity - in some languages additional stress tends to be placed on syllables that are longer (moraically heavy).
Prosodic stress is also often used pragmatically to emphasize (focus attention on) particular words or the ideas associated with them. Doing this can change or clarify the meaning of a sentence; for example:
I didn't take the test yesterday. (Somebody else did.)
I didn't take the test yesterday. (I did not take it.)
I didn't take the test yesterday. (I did something else with it.)
I didn't take the test yesterday. (I took one of several. or I didn't take the specific test that would have been implied.)
I didn't take the test yesterday. (I took something else.)
I didn't take the test yesterday. (I took it some other day.)
As in the examples above, stress is normally transcribed as italics in printed text or underlining in handwriting.
In English, stress is most dramatically realized on focused or accented words. For instance, consider the dialogue
"Is it brunch tomorrow?"
"No, it's dinner tomorrow."
In it, the stress-related acoustic differences between the syllables of "tomorrow" would be small compared to the differences between the syllables of "dinner", the emphasized word. In these emphasized words, stressed syllables such as "din" in "dinner" are louder and longer. They may also have a different fundamental frequency, or other properties.
The main stress within a sentence, often found on the last stressed word, is called the nuclear stress.
In many languages, such as Russian and English, vowel reduction may occur when a vowel changes from a stressed to an unstressed position. In English, unstressed vowels may reduce to schwa-like vowels, though the details vary with dialect (see Stress and vowel reduction in English). The effect may be dependent on lexical stress (for example, the unstressed first syllable of the word photographer contains a schwa , whereas the stressed first syllable of photograph does not /'fo?tgræf -gr?:f/), or on prosodic stress (for example, the word of is pronounced with a schwa when it is unstressed within a sentence, but not when it is stressed).
Many other languages, such as Finnish and the mainstream dialects of Spanish, do not have unstressed vowel reduction; in these languages vowels in unstressed syllables have nearly the same quality as those in stressed syllables.
Some languages, such as English, are said to be stress-timed languages; that is, stressed syllables appear at a roughly constant rate and non-stressed syllables are shortened to accommodate that, which contrasts with languages that have syllable timing (e.g. Spanish) or mora timing (e.g. Japanese), whose syllables or moras are spoken at a roughly constant rate regardless of stress. For details, see Isochrony.
It is common for stressed and unstressed syllables to behave differently as a language evolves. For example, in the Romance languages, the original Latin short vowels /e/ and /o/ have often become diphthongs when stressed. Since stress takes part in verb conjugation, that has produced verbs with vowel alternation in the Romance languages. For example, the Spanish verb volver has the form volví in the past tense but vuelvo in the present tense (see Spanish irregular verbs). Italian shows the same phenomenon but with /o/ alternating with /uo/ instead. That behavior is not confined to verbs; note for example Spanish viento "wind" from Latin ventum, or Italian fuoco "fire" from Latin focum.
An operational definition of word stress may be provided by the stress "deafness" paradigm. The idea is that if listeners perform poorly on reproducing the presentation order of series of stimuli that minimally differ in the position of phonetic prominence (e.g. [númi]/[numí]), the language does not have word stress. The task involves a reproduction of the order of stimuli as a sequence of key strokes, whereby key "1" is associated with one stress location (e.g. [númi]) and key "2" with the other (e.g. [numí]). A trial may be from 2 to 6 stimuli in length. Thus, the order [númi-númi-numí-númi] is to be reproduced as "1121". It was found that listeners whose native language was French performed significantly worse than Spanish listeners in reproducing the stress patterns by key strokes. The explanation is that Spanish has lexically contrastive stress, as evidenced by the minimal pairs like tópo ("mole") and topó ("[he/she/it] met"), while in French, stress does not convey lexical information and there is no equivalent of stress minimal pairs as in Spanish.
An important case of stress "deafness" relates to Persian. The language has generally been described as having contrastive word stress or accent as evidenced by numerous stem and stem-clitic minimal pairs such as /m?hi/ [m?.hí] ("fish") and /m?h-i/ [m.hi] ("some month"). The authors argue that the reason that Persian listeners are stress "deaf" is that their accent locations arise postlexically. Persian thus lacks stress in the strict sense.
The orthographies of some languages include devices for indicating the position of lexical stress. Some examples are listed below:
Though not part of normal orthography, a number of devices exist that are used by linguists and others to indicate the position of stress (and syllabification in some cases) when it is desirable to do so. Some of these are listed here.
') before the stressed element, secondary stress by a low vertical line (secondary stress mark:
?). For example, [slæb?f?'ken] or /slæb?f?'ken/. Extra stress can be indicated by doubling the symbol: ''?.