Connected speech, as described by Gimson (2008), is “an utterance consisting of more than one word” (p. 263). There are several features which characterize the connected speech and which need to be taken into consideration to achieve a good level of pronunciation. It is not enough to master individual phonemes of a language but on the other hand, to sound as correct and intelligible as possible it is necessary to know what the phonetic aspects are when phonemes are combined into words and words into sentences. These phonetic features known as suprasegmental can be defined as the features which “stretch over more than a single segment – possibly a syllable, a complete word or phrase, whole sentences, or even more” (Collins & Mees, 2008, p. 124). They include rhythm, stress, assimilation, liaison, elision, intonation etc.
Skaličková (1974) in her work cites several phoneticians who emphasize the importance of the aspects in connected speech as for example J. D. O'Connorin his work Stress, Rhythm and Intonation argues, “Many students learn to make the individual sounds correctly enough, yet their speech remains barely intelligible to the English ear. The reason for this paradox is usually to be found in faulty rhythm and intonation” (p. 124). Or W. S. Allen who in his Living English Speech states that “a reasonably correct speech-flow is more important for intelligibility than correct sounds” (as cited in Skaličková, 1974, p. 124).
Some of the suprasegmental features are discussed more often in various works on phonetics while some of them are considered to be less important. This work focuses only on some of the above mentioned aspects which seem most relevant for the work.
Stress as defined in the Oxford Dictionaries is “an emphasis given to a particular syllable or word in speech, typically through a combination of relatively greater loudness, higher pitch, and longer duration” (“stress,” 2015). Gimson (2008) besides these three factors states also the forth one – quality of the vowels (p. 236). All these four factors cause certain parts or syllables to be more prominent; however, some of them seem to have more significance. As stated by Gimson (2008) “it is principally pitch change which marks an accented syllable” (p. 236). Also Roach (1991) in his work asserts that “these factors are not equally important; the strongest effect is produced by pitch, and length is also a powerful factor. Loudness and quality have much less effect” (p. 86).
Pitch assigns the degree of highness of sounds as perceived by a human. The most important factor which governs the quality of the sounds is the rate of vibration or frequency of the vocal folds (Collins & Mees, 2008, p. 133). Collins and Mees (2008) state that “the higher the frequency, the higher the perceived pitch” (pp. 124, 133) and the higher the pitch, the stronger the stress.
The other factor which on the part of the listener is loudness would be intensity. The level of intensity depends on how much breath effort and muscular energy are expended by the speaker. The greater effort and energy means the stronger intensity and that associates with the stressed syllables (Collins & Mees, 2008, p. 124), although, as mentioned before, loudness is not so significant in determining the stress in English (Gimson, 2008, p. 237).
Vowel quality is highly connected to so-called vowel reduction in the context of unstressed syllables. Vowel reduction is realized when “the peripheral vowel in the unstressed syllable is actually replaced by another phoneme – most commonly by /ə/, sometimes by /ɪ/ or /ʊ/, or even a syllabic consonant, e.g. attention [əˈtenʃn]” (Collins & Mees, 2008, p. 124). The quality of the vowels in a word changes if the placement of the stress changes. When compared the noun present[ˈprezənt] to the verb (to) present [prəˈzent] the peripheral vowel /e/ can be found in the first syllable of the noun but in the second syllable of the verb. The opposite can be seen with the central vowel /ə/. This means that the peripheral vowel /e/ occurs in the stressed syllables while the central vowel /ə/ occurs in the unstressed syllables instead. Thus, vowel reduction causes the unstressed syllables to be less prominent (Collins & Mees, 2008, p. 124).
When the quantity (duration) of vowels is considered, vowels are longer in stressed than in unstressed syllables. It can be noticed in an example by Collins and Mees (2008) when compared the length of the vowel in two words with different stresses: sarcasm[ˈsɑːkæzəm] and sarcastic [sɑˈkæːstɪk] (p. 125). In spite of the fact Gimson (2008) states, “Despite the lesser prominence of all short vowels, a long vowel in an unaccented syllable is sometimes longer than a short vowel in an adjacent accented syllable, e.g. pillow [ˈpɪləʊ], record [ˈrekɔːd]” (p. 237).
Collins and Mees (2008) distinguish between word stress and sentence stress (p. 124). The latter one is meant as the stress in connected speech involving the subject of strong and weak forms.
220.127.116.11. Word stress
In English each individual word carries its own stress when it stands in isolation. In contrast to some languages which have regular stress pattern, English stress can occur in different positions depending on each individual word. Gimson (2008) claims the stress pattern to be both fixed and free. With certain exceptions stress has always its particular place in any given word and therefore it is fixed, but it is free when taken into consideration that in general there is no particular syllable on which stress falls (p. 235).
There can be more levels of stress. In some words besides the main stress (also called primary or principal stress) occurs a syllable which is not as strongly stressed as the one with the main stress but still carries more prominence than the unstressed syllable. This is assigned as a secondary stress. For example, the word photographic would be transcribed as [ˌfəʊtəˈgræfɪk] where the first syllable carries the secondary stress, represented by a low vertical line, while the primary stress is as usually indicated by a vertical line at the top just before the stressed syllable (Roach, 1991, p. 87).
Although there exist some rules on English stress patterns, they involve too many exceptions and so it is very hard for a foreign learner to predict the main stress of the words just from the written form. The rules are very complex and this thesis does not focus on them for the sake of the given extent and also for the sake of complexity and beliefs of certain phoneticians that it is better for the foreigners to learn the stresses by heart together with individual words (Roach, 1991, p. 88).
18.104.22.168. Sentence stress
As cited in Kingdon’s definition on sentence stress, it is “the relative degree of force given to the different words in a sentence” (as cited in Pavlík, 2000, p. 181). The degree of the force depends on how important the word in the sentence is i.e. how much information it conveys. Generally the more information the word carries the more stressed it becomes. Based on this general rule, the words which convey only little information often lose their stresses in connected speech. (Collins & Mees, 2008, p. 130; Pavlík, 2000, p. 181). These are usually function words which play important structural role for the sentence but their lexical meaning is almost none. These involve articles, prepositions, pronouns, auxiliary verbs, conjunctions, determiners and some adverbs. Function words have their own stress when they stand isolated in which case they are said to be in their strong form. However, they usually appear as the parts of sentences in their weak unaccented forms (Pavlík, 2000, p. 173). By contrast there are content words such as nouns, adjectives, main verbs, numerals and most adverbs in which “lexical meaning prevails over their grammatical meaning” (Pavlík, 2000, p. 182). The difference can be seen on the example taken from Collins and Mees (2008, p. 130):
I’ve ˈheard that ˈJack and ˈJane ˈspent their ˈholidays in Jaˈmaica.
F F C F C F C C F C F C
(C = content word, F = function word)
When the function word is being used in its weak form, several phenomena can take place, involving (Pavlík, 2000, pp. 173-174):
There exist some cases when function words are used in their strong forms in connected speech. This applies mainly to (Collins & Mees, 2008, p. 130):
(a) wh-words where these form questions, e.g. where, why, how
(b) demonstratives, e.g. this, that, these, those
Function words indicating a contrast:
I said give it to ˈhim, not ˈher.
Collins and Mees (2008) also talk about the case when some of the content words become unstressed which happens if the utterance is said at more rapid tempo (p. 130):
I'veheard that ˈJack and ˈJane spent their ˈholidays in Jaˈmaica.
The use of the weak forms in English is very frequent and every learner of the English language and its pronunciation should practise this phenomenon. The reason for this is not only to achieve sounding more natural and more native-like but to be aware of the weak forms means also better understanding of the speakers who use them (Pavlík, 2000, p. 73).
The feature which gives English an impression of being rhythmical and which is the basis for the rhythm in English is sentence stress. English is assigned to a group of languages called the stress-time languages (Collins & Mees, 2008, p. 131). Pavlík (2000) in his work states the definition by Crystal which says, “In stress-timed languages, it is claimed that the stressed syllables recur at regular intervals of time, regardless of the number of intervening unstressed syllables” (p. 186). Of course the regularity is relative and the time period occurring between the stressed syllables is not exactly the same but English has “a tendency towards taking an approximately equal period of time between one stressed syllable and the next” (Pavlík, 2000, p. 187).
Such stress-timing is related to so called borrowing rule which is about shortening some vowels due to the others. Gimson (2008) defines it as the rule of English rhythm “whereby a syllable with a reduced vowel ‘borrows time’ from any immediately preceding syllable containing a full vowel” (p. 265). In this way the stressed syllables followed by reduced syllables become shortened while the other stressed syllables are equally long and reduced syllables are equally short (Gimson, 2008, p. 265). The example by Collins and Mees (2008) shows the shortening of the vowel in a word as unstressed syllables are added (p. 131):