15 Morpho-phonemics
Mr. Pankaj Dwivedi
Learning Outcomes:
After the completion of this module, the students would gain an insight into the key concepts of English morphophonemics. Students would be able to understand how morphological structure of a morpheme may affect pronunciation or phonological shape of following or preceding morphemes.
Introduction
This chapter presents an introduction to English morphophonemics. Sound structure (phonology) and word structure (morphology) are two of the main components of a language system. However, many a times, these systems interact and affect each other in some interesting ways and give rise to a phenomenon which is known as ‘morphophonemics’ Sometimes, due to this interaction, pronunciation of a morpheme may get modified or completely changed. These changes may be regular or irregular and usually are context sensitive in nature. In ‘morphophonemics’, we specifically study the changes which occur at the margins/boundaries of morphemes. There are many glossaries, lectures, notes and other reference materials available over the internet for further studies in this area. Some of the pioneer reference works are also listed in the reference section.
Note: In this module, IPA transcriptions of morphemes/words have been throughout referred from Cambridge Online dictionary.
Background
In his book titled ‘A Manual Of Phonology’ Charles F. Hockett, Professor of linguistics and anthropology, says that language is nothing but a complex system of habits and this system can be broken down and categorized into five principle subsystems:
1. The phonological system: refers to the stock of phonemes and the systematic arrangements in which these phonemes occur in a language;
2. The grammatical system: refers to the stock of morphemes, the systematic arrangements in which these morphemes occur in a language;
3. The morphophonemic system: refers to the relational code that ties phonological and grammatical system together.
4. The semantic system: refers to the association of morphemes, combinations of morphemes, and the systematic arrangements in which morphemes can be put, with things and situations, or kinds of things and situations.
5. The phonetic system: refers to the process in which phonemes, sequences of phonemes of are converted into sound waves/speech signals by a speaker, and these sound waves/speech are then decoded from by a hearer.
Of these five principle subsystems, 1 to 3 are central, while 4-5 seem to be peripheral in nature to the language as a system. In this chapter, our main focus will be on 3 ‘The morphophonemic system’ and this focus will further be limited to scope in that examples and discussions will be revolve around morphophonemic system of English language.
Morphophonemics
Morphophonemics may be defined as analysis and classification of the phonological factors which affect the pronunciation of morphemes or, correspondingly, the morphological factors which affect the appearance of phonemes. In morphophonemics, we basically study interaction between morphological and phonological processes and how they these factors affect each other. Morphophonemic change usually occurs at morpheme boundaries and it involves sounds that are associated with separate phonemes.
One very obvious example to morphophonemics would be the use of indefinite in English language. Indefinite article in English has two manifestations: a and an. If a word begins with a consonantal sound then indefinite article is manifested as ‘a’ (a mango or a cat), while it is manifested as ‘an’ (an apple or an idiot) if following word starts with a vowel sound.
Note: The term ‘sound’ refers to the way a phone (consonant or vowel) is pronounced, not necessarily written, in English. Therefore, an hour is a correct phrase not *a hour and a university is the correct phrase not *an university.
Essential Vocabulary
To understand these kind of sound changes, i.e., morphophonemics, it’s important to give a quick look to following concepts: the smallest unit of sound (both underlying and surface forms – phoneme and phone, respectively) minimal unit of meaning (both underlying and surface forms – morpheme and morph, respectively), and how these phonemes and morphemes may show variation in their shape given different contexts – known allophony and allomorphy, respectively). Allomorphs are derived from a single morpheme known as underlying representation with the help of morphophonemic rules.
A. Phoneme: a phoneme is considered as a smallest unit of human speech sound that distinguishes meaning. Phonemes are usually written within slanted brackets //. There are 26 letters in English alphabet. But these 26 letters represent 44 phonemes. For example, phoneme /k/ occurs in words such as cut, kite, school and skin.
B. Phone: a phone is a surface manifestation/form of a phoneme given a phonological environment. Phones are usually written within square brackets []. In English, phoneme /p/ is manifested (pronounced) as [p] in words like spat and spoon but as [ph] in words like pin, pick and poll.
C. Morpheme: a morpheme is a minimal, i.e. indivisible, unit of meaning. Morphemes are of various types given their structure (free and bound), content (zero and null) and functions (derivational and inflexional).
D. Morph: a morph is a surface manifestation/form of morpheme given a phonological environment. For example, plural morpheme is realized as [s] (cats and bats), [z] (dogs and calls) and [əz] (judges and churches) given different phonological environment.
E. Allophone: an allophone is a phonetic variant of a phoneme in a particular language.
F. Allomorph: an allomorph is one of two or more complementary morphs which manifest a morph in its different phonological or morphological environments.
G. Underlying Representation: an underlying representation is the most basic form of a morpheme/word before any phonological, morphological or morphophonological rule has been applied to it. Underlying representation shows native speaker’s knowledge about the abstract underlying phonology of the language.
H. Surface Representation: A surface representation (actual occurrence) is the form of a word that is spoken and heard.
I. Morphophonemic rules: A morphophonemic rule is written in form of a phonological rule but it is restricted to a particular morphological environment. Unlike phonological rules, morphophonemic rules are sensitive to their environment. Whenever morphological information is required to specify the environment for an phonological rule, the rule is morphophonemic.
J. Phonological conditioning: phonological conditioning refers to the phenomenon wherein phonological environment of a morpheme determines selection and shape of its allomorph(s).
K. Morphological conditioning: morphological conditioning is a conditioning wherein morphological environment of a morpheme determines selection and shape its allomorph(s).
Steps involved in Morphophonemic Analysis
Morphemes are stored in lexicon. They are strung together by morphological and syntactic rules. These morphemes are then converted to their surface forms by a sequence of ordered phonological rules. In doing morphophonemic analysis, our aim is to establish a transparent connection – as far as it is possible – between data and theory. The main purpose of morphophonemic analysis is to discover a set of underlying forms and the ordered rules which are consistent with the given data. Main steps involved in the process of morphophonemic analysis are as following:
I. Phonemicization: It is easier to perform morphophonemic analysis with data that are already expressed as phonemes. Therefore, it is recommended first to reduce the data into phonemes. This step is also referred as pre-processing of data in the process morphophonemic analysis.
II. Morpheme division: The second step is to divide the words forms into their component morphemes. However, phonological alternations tend to obscure this division.
III. Finding the allomorphs: Keeping visible and possible phonological alternation in mind, it’s necessary to find the allomorph of each alternating morpheme. After allomorphs have been indentified, find the segment(s) which tend to alternate.
IV. Active rules: After division of words into morphemes and identification of allomorphs, it is possible to state rules and their order of application.
V. Setting up underlying representations: Taking into account the logical possibilities, underlying representations are to be set up so that all the possible allomorphs of each morpheme can be derived from a single underlying representation using general phonological rules. However, choosing underlying representations often involves considering more than one hypothesis. The following steps are usually helpful in determining correct underlying form:-
- Segments exhibiting X∼Y alternation are underlyingly /X/, which is converted to [Y] in certain contexts by one or more phonological rules.
- Segments exhibiting X∼Y alternation are underlyingly /Y/, which is converted to [X] in certain contexts by one or more phonological rules.
- Surface segments that do not alternate are assumed to be the same at the underlying level.
- For segments that alternate, consistently use the underlying segment. So, if there was a X∼Y alternation and you determined that underlyingly the segment was /X/, then consistently write /X/ as the underlying representation in all lexcial items with the X∼Y alternation.
- Each morpheme has only one underlying representation form.
Note: An underlying form does not always necessarily correspond to the pronunciation of some pronounced word.
Morphophonemics Processes:
Pluralisation in English
To understand the morphophonemics, it’s very important to understand concepts A to K not only in isolation but also as to how they appear in a morphophonemic process. Here, we will present regular pluralisation process of English language. Below are given four set of single morpheme words and their plural counterparts in English. Both singular and plural forms have been presented with their respective IPA transcriptions.
Table 1
| Word (Singular) | IPA Transcription | Word (Plural) | IPA Transcription |
| cut | /kʌt/ | cuts | /kʌts/ |
| kick | /kɪk/ | kicks | /kɪks/ |
| top | /tɒp/ | tops | /tɒps/ |
| dog | /dɒɡ/ | dog | /dɒɡz/ |
| rib | /rɪb/ | rib | /rɪbz/ |
| kid | /kɪd/ | kid | /kɪdz/ |
| judge | /dʒʌdʒ/ | judges | /dʒʌdʒəz/ |
| wish | /wɪʃ/ | wishes | /wɪʃəz/ |
| church | /tʃɜːtʃ/ | churches | /tʃɜːtʃəz/ |
| shoe | /ʃuː/ | shoes | /ʃuːz/ |
| day | /deɪ/ | days | /deɪz/ |
| pea | /piː/ | peas | /piːz/ |
Upon careful observation of the data, it is noticed that plural morpheme in English is realized as /-s/, /-z/ and /-əz/ given different phonological environment of the stem to which plural morpheme gets attached to. When one morpheme takes more than one form (morph) in different phonological or morphological environment, these morphs are referred as allomorph to one another. In this kind of situation, we need to set up a single underlying representation from which other morphs are derived for their pronunciation in any particular context. This underlying representation is achieved with the help of some rules, usually referred as morphophonemic rules.
To see how it works in English, we will consider the phonological environment of all three sets. Since in English language, plural morpheme gets attached to the right boundary (word+plural morpheme) of a morpheme, phonological environment of the right boundary will be taken into account for the consideration.
- First set consisting of ‘cut, kick and top’ morphemes have their right boundary ending in voiceless consonant which is non-strident. These morphemes take /-s/ allomorph.
- Second set consisting of ‘dog, rib and kid’ morphemes have their right boundary ending in voiced consonant which is non-strident. These morphemes take /-z/ allomorph.
- Third set consisting of ‘judge, wish and church’ morphemes have their right boundary ending in a consonant that strident in nature. These morphemes take /-əz/ allomorph.
- Fourth set consisting ‘shoe, day and pea’ morphemes have their right boundary ending in vowel. These morphemes also take /-z/ allomorph.
Therefore, it is evident allocation of plural allomorph in English is not random in nature but is conditioned by phonological environment occurring at right boundary of the morpheme. Usually, the allomorph with the wisest distribution qualifies as a suitable candidate for underlying representation. In this case, /-z/ qualifies to become underlying allomorph as it occurs after most voiced consonants and after all vowels. And from allomorph /-z/, other allomorphs /-s/ and /-əz/ and //have been derived. This whole process can be explained by (a set of) rules known as morphophonemic rules.
Table – 2
| word + plural morpheme | cut-s | dog-s | judge-s | order of rule application |
| underlying representation | /kʌt-z/ | /dɒɡ-z/ | /dʒʌdʒ-z/ | |
| coda epenthesis | – | – | /dʒʌdʒəz/ | |
| devoicing | /kʌts/ | – | – | |
| surface form | [kʌts] | [dɒɡz] | [dʒʌdʒəz] |
It is to be noticed that it is not the correct morphophonemic rules but also the correct ordering of these rules is equally crucial. In absence of correct ordering, produced surface forms may well be incorrect. For example, if we change the ordering of coda epenthesis and devoicing rules in the above example, /dʒʌdʒs/* will get produced as a surface form for judges.
Table -3
| word + plural morpheme | cut-s | dog-s | judge-s | order of rule application |
| underlying representation | /kʌt-z/ | /dɒɡ-z/ | /dʒʌdʒ-z/ | |
| devoicing | /kʌts/ | – | dʒʌdʒ-s/ | |
| coda epenthesis | – | – | – | |
| surface form | [kʌts] | [dɒɡz] | [dʒʌdʒs]* |
It is clear that pronunciation, i.e., surface appearance, of plural allomorph in English depends on the phonological environment occurring at the right boundary of the morpheme. This phenomenon is also known as phonological conditioning and allomorphs /-s/, /-z/ and /- əz/ are also referred as phonologically conditioned allomorphs. However, the regularity of phonological conditioning is restricted in its scope, that is, several morphemes despite of having suitable phonological environment for undergoing change may exhibit no or some different kind of phonological change, i.e. the proximity of a sound(s) does not affect these forms. For example, –en as a plural making morpheme is a peculiar case to only certain words such as children, oxen, brethren, etc. These kinds of changes are to due to what is called morphological conditioning. Occurrence of a morph/form is determined by the morpheme it gets attached to rather than its phonological environment. To further support the concept, we will present another subclass of plurals in English, which defies the present phonological environment and does not let the morpheme undergo any change whereas another set having same phonological environment undergoes predictable changes.
English has a restricted class of words in which sound /f/ at the right margin of the stem/base alternates with /v/ when there is a plural suffix gets attached to them. However, the alternating class (first two columns) is unproductive. Any newly coined word(s) with final /f/ entering in English does not exhibit the f-v alternation.
Table – 4
|
Alternating class
|
Non-alternating class | ||
| Singular | Plural | Singular | Plural |
| thief | thieves | chief | chiefs |
| wolf | wolves | laugh | laughs |
| knife | knives | fife | fifes |
| leaf | leaves | whiff | whiffs |
The process responsible for this f-v alternation and their respective derivations from underlying form /z/ is given as follows (Table 5):
/f/ Voicing
f → v / ]-Plural
Table – 5
| Word (alternating
class) |
Plural form | Word (non-alternating
class) |
Plural form | order of rule application
|
|
| plural forms | knife | knives | laughs | laughs | |
| underlying | /naɪf/ | /naɪf-z/ | /lɑːf/ | /lɑːf – z/ | |
| representation | |||||
| f → v rule | /naɪvz/ | ||||
| devoicing | /lɑːfs/ | ||||
| Surface form | [naɪvz] | [lɑːfs] |
As it is clear from the Table-3, order of rule application is equally important in this case too. Else, we will end up deriving incorrect plural form /naɪvs/*.
After discussing pluralization process, we can move to some other morphophonemic changes which are commonly found in English and other languages of the world.
Assimilation of Negative prefix ‘in-’ in English
Assimilation is a sound change in which one sound becomes more like a following sound. This may occur either within words/morphemes (phonological change) or between words/morphemes (morphophonemic change). In the prefixation process, final sound (marginal sound) of a prefix becomes more like the first sound of a base/root to which it attaches to. However, it usually happens when the articulation of the two sounds – last sound of the prefix and first sound of the base – carries some degree of feature similarity. The assimilation, however, does not always necessarily occur between two sequential sounds. Sometimes, sounds which are in distance, i.e., having some other sounds in-between, also tend to assimilate.
To understand how assimilation affects pronunciation in English, we will examine the below given data:
Table – 6
| Word | Negative prefix + word | IPA Transcription |
| possible | in+possible | /ɪmposɪbəl/ |
| balance | in+balance | /ɪmbæləns/ |
| tolerable | in+tolerable | /ɪntɒlrəbl/ |
| decent | in+decent | /ɪndiːsənt/ |
| active | in+active | /ɪnæktɪv/ |
| alienable | in+ alienable | /ɪnelɪənəbl/ |
| complete | in+ complete | /iŋkʌmpliːt/ |
| ɡratitude | in+ɡratitude | /ɪŋɡrætɪtjuːd/ |
- Allomorph [im-] occurs only before when following consonant is labial in nature–[e.g. p, b, f, m].
- Occurrence of Allomorph [ɪŋ] only before the velar consonants – [k , g]
- Elsewhere, allomorph [in] is found to be appearing, that is to say, before alveolar consonants such as [t, d, s, z, n] and before vowels sounds such as [ə, ʊ, u, ɪ, iː]
While negative prefix in-’ shows a tendency for assimilating with the place of articulation of the following sound, the other negative prefixes such as ‘un-’ and ‘non-’ do not show any regular characteristics of assimilating behaviour. See some examples below:
un- + polluted → unpolluted (*umpolluted)
non- + personal → non-personal (*nom-personal)
un- + balanced → unbalanced (*umbalanced)
non- + business → non-business (*nom-business)
( un- + marked → unmarked (*ummarked)
non- +medical → non-medical (*nom-medical)
(un- + lawful → unlawful (*ullawful)
non- + legal → non-legal (*nol-legal)
un- + reasonable → unreasonable (*urreasonable)
non- + restrictive → non-restrictive (*nor-restrictive)
[see Okada, 2013
If in- could be attached to all the words above instead of un- and non-, assimilation would be expected to take place.
Shift in English:
Of the several active morphophonemic processes in English, one is shifting of stress from one syllable to another whenever a new morpheme gets attached to it. English is stress timed language and primary stress falls normally on the first syllable of the word. However, it may appear on any syllable. Usually, deriving words via affixation may lead to shift of stress and thereby may lead to modification or change in one or more vowels. This is also addressed as vowel gradation or vowel alternation process. Let us have a quick look to the following data given Table 7. Syllable carrying primary stress have been marked in bold case.
Table -7
| word | 1st derivation | 2nd derivation | 3rd Derivation |
| photo | photograph | photography | photographic |
| /ˈfəʊ.təʊ/ | /ˈfəʊ.tə.ɡrɑːf/ | /fəˈtɒɡ.rə.fi/ | /ˌfəʊ.təˈɡræf.ɪk/ |
| drama | dramatic | dramatically | |
| /ˈdrɑː.mə/ | /drəˈmæt.ɪk/ | /drəˈmæt.ɪ.kl.i/ |
As it is clear from the Table-7 that affixation process may lead to the shift of stress or gradation of vowels, however, this kind of change may be regular with one set of affixes and completely sporadic with other ones. Hence, rules describing stress patterns in English are not exhaustive in nature. Some rules may, anyway, be stated for the common understanding:-
Grammatical Conditioning andSuppletion:
Unlike what we majorly discussed in the previous sections (Pluralisation in English and assimilation of negative prefix in-), sometimes phonological factors play no role in the selection of a particular allomorph. It is rather determined by a certain grammatical class. Its choice may be dependent on a certain grammatical class. A special allomorph may be required in a given grammatical context; there may not be any good phonological reason for its selection, however. For example, in irregular verbs such as see in English. see, saw, seen determined by the present tense, past tense, and the non-progressive participle (these are grammatical features). There also exist a few morphemes whose allomorphs show no phonetic similarity. For example, comparative adjective form of good, i.e. better, contains lexeme good but there is not a single common sound found between both of them. These kinds of cases, wherein two morphemes are phonetically unrelated, are referred as cases of suppletion.
Useful notations
- Since morphemes are made of phonemes, they are expressed between slashes ‘// ’.
- The boundary between two morphemes is often represented either with plus sign ‘+’. Sometimes linguists also use a hyphen ‘-’ for expressing morpheme boundary.
- Word boundaries are usually indicated with hash sign ‘#’.
- A rule A B / C _ D, will be read as A changes into B in the environment when C precedes the sound A and D follows the sound A. The forward arrow ( ) translates into ‘changes into’ and forward slash (/) translates into ‘in the environment of’. A sometimes is also addressed as target, B as change and C & D are trigger.
- Notation ‘_’ is often referred as focus bar and this is where the change takes place.
- Symbol ‘Ø’ for expressing process of deletion and epenthesis.
- σ – labelled parenthesis is used to indicate the syllable boundary.
- Asterisk (*) refers to ungrammatical or unacceptable string of words.
Summary
Phonology is regarded as the foundation for other branches of linguistics. Morpheme, a minimal distinctive unit of meaning, constitutes of nothing but a set of phoneme(s), a minimal unit of sound. The phonemes tend to behave interestingly when they get combined to produce a morpheme or other longer units such as syllables, words, phrases and sentences. Their combine (phonological) effect often crosses over so called boundaries in several subtle and visible ways. Sometimes the effect could be synchronic and sometimes it may be diachronic in nature. These effects may be regular or completely sporadic in their occurrences. In the discipline of morphophonemics, we study as to how morphological and phonological factors affect each other usually at morpheme boundaries involving separate phonemes. Morphophonemic properties of English or any other language for that matter are of great importance in order to analyze other linguistic properties of that language.
| you can view video on Morpho-phonemics | |
Reference
- Chomsky, N., & Halle, M. (1968). The sound pattern of English. MIT Press.
- Chomsky, N., & Halle, M. (1960). The morphophonemics of English. MIT Press.
- Goldsmith, John A, et. al. ed (2014).The Handbook of Phonological Theory Oxford: Blackwell.
- Hayes, B. (2011). Introductory phonology (Vol. 32). John Wiley & Sons.
- Kilbury, J. (1976). The development of morphophonemic theory (Vol. 10). John Benjamins Publishing.
- Katamba, F., & Stonham, J. (1993). Morphology (Modern Linguistics Series). London: Tottenham Court Road, 19-20.
- Okada, A. (2013). Phonological Features of in-, un-and non. International Proceedings of Economics Development and Research, 68, 123.
- Loos, Eugene E., et. al. ed. (2004). Glossary of Linguistic Terms: SIL International (retrieved from http://www-01.sil.org/linguistics/glossaryoflinguisticterms)
- Pike, K. L. (1971). Phonemics: a technique for reducing languages to writing. Michigan University Press
- Spencer, A. (1995). Phonology: theory and description (Vol. 9). Wiley-Blackwell.