The Analysis on Segmental and Supra-Segmental Features in

Robert Frost’s Poem Titled Fire and Ice

Octovianus Bin Rojak
octovianusdb07@gmail.com
Sanata Dharma University

Abstract
This paper is a study of Robert Frost poem with the title Fire and Ice. The aim of this
study is to analyse the segmental and supra-segmental features in the poem and use it to
help readers familiar with those features which later help in mastering the spoken
language. This study is the qualitative study where the main data used in this study is the
poem composed by Robert Frost the retrieved from Poetry Foundation’s website. This
study discusses the segmental aspects consist of consonants and vowels and supra-
segmental aspects consist of stress, intonation, and rhythm. The knowledge of both
segmental and supra-segmental aspects is really important to analyse and recite the poem
where later helping readers to master the spoken language through intensive exposure to
those aspects.

Keywords: Robert Frost, poem, segmental, supra-segmental

INTRODUCTION

As a second language in Indonesia, English only ha developed recently due to the

urgent need in this global era. Starting from a job application, usually, one of the
requirement is mastery of English or usually written in the job advertisement as fluent in
oral and written English. It is important especially for a multinational company when the
colleagues will come from different countries and the only way to communicate with
them is using English language as the vehicle of communication. Another example the
important need in English is for an Indonesian student who wishes to study abroad, the
campus he/she want to enrol require him/he to give a proof of English language
proficiency which can be earned by taking English proficiency tests like IELTS or
TOEFL IBT.
Unfortunately, most Indonesians still not reach that level of proficiency. It is due to
the exposure of that language. English language is only taught in formal school starting
from junior high school and it is quite late for Indonesians to acquire the language. Apart
from that, Indonesia also has a deficiency in the number of competent English teachers in
formal school. In teaching and learning process in school, problem arises since the
language only treated as an object of intellectual study where students are being taught
just to memorise things rather than understanding it. After understanding it, later the
students can acquire it well. Acquiring the language means they mastered both the written
and spoken language.
The very obvious measurement to measure whether someone has good proficiency
in English is through the spoken language (Alip, 2016). In his paper, he introduced the
segmental and supra-segmental aspects of English phonology to Indonesian students as an
effort to make the students and teachers of English aware of the importance of mastery
the spoken English language.
As suggested in his paper that in accordance with the title of this paper, to treat
segmental and supra-segmental errors in spoken English are through exposure of native
pronunciation and systematic minimal pairs, and through the exposure of native English
utterances, respectively, thi s paper is written to analyse the segmental and
supra-
segmental features in Robert Frost’s poem titled Fire and Ice.

LITERATURE REVIEW

This paper used two theories as for the basis namely the segmental features consist
of consonant and vowel phonemes, and supra-segmental features comprise of stress,
rhythm, and intonation.

A. Segmental
Phonemes are the minimal units or one of a small speech sounds or also known as
segments that are distinguished by the speakers of a particular language to represent
differences in meaning. Using substitution test, the phoneme status can be proven through
the substituting one segment for another that can produce a different word, for example,
the phoneme of /e/ and /ɪ/ in desk and disk. Any pair of words exists that differs in one
sound only or distinguished by just one segment called a minimal pair (Poole, 1999: 57).
In English, the segments of phonemes consist of two aspects namely the consonant
phonemes and vowel phonemes.

A.1 Consonant Phonemes

In English, there are 24 consonant phonemes which can be categorised into 6 oral
stops {/p, b, t, d, k, g/}, 3 nasal stops {/m, n, ŋ}, 4 approximants {/l, r, j, w/}, 2 affricates
{/tʃ, dʒ/}, and 9 fricatives {/f, v, ð, θ, s, z, ʃ, ʒ, m, h/} (Giegerich, 1992: 41). Comparing
the English consonant phonemes with Indonesian consonants phonemes, there are slight
differences. The most differences found in fricatives and affricates. For Indonesian, they
perceived the phonemes /f/ and /v/ as /p/ while /ʃ/, /z/, /ʒ/ as /s/. The latter is both
phonemes are alveo-palatal in Indonesia while in English the phonemes are palato-
alveolar or also labelled as postalveolar (Alip, 2016: 16).

A.2 Vowel Phonemes

The vowel in English, specifically in Received Pronunciation (RP) consists of 12
simple vowel phonemes and 8 diphthongs (Poole 1999: 60), the same with General
American (GA) (Ladefoged 1993: 31). The quality of the vowel can be distinguished
between lax (short) consist of 6 vowel phonemes and tense (long) consist of 10 phonemes.
The lax vowels are {/ɪ, e, æ, ʊ, ʌ, ɝ/} and the tense vowels are {/i, eɪ, ɑ, ɔ, oʊ, u, ju, aɪ, aʊ,
ɔɪ /} (Ladefoged 1993: 87). There is a vowel known as schwa /ə/, used to designate
vowels that have reduced vowel quality or the vowel is unstressed during the realisation
in a connected speech which will be discussed further in the discussion section.
Comparing the number of phonemes and the quality of vowels, it is difficult for
Indonesian to pronounce or speak the English words since in Indonesian; there are only
six vowel phonemes and they do not have tense vowels since it is not phonemic (Alip,
2016: 16).

B. Supra-segmental
Supra-segmental is the units of linguistic which operate above single sounds (Alip,
2016: 16). The supra-segmental features consist of stress, rhythm, and intonation (Collins
and Mees, 2013:128) that will be explained further below.

B.1 Stress
When there is a word in English consists of more than one syllable or known as
polysyllabic, certain syllable will receive a higher degree of prominence or stressed.
Stress syllables are produced with a stronger eruption in initiatory energy and the energy
result in greater loudness (Giegerich, 1992: 179). In addition, stress also phonemic in
English that is to say that different stress will produce different meaning. For example,
the word pervert can have two meanings one is a verb and one is a noun. If pronounced as
ˈpervert, where the stress is on the first syllable, it means that the word is noun that has
meaning of a person whose behaviour deviates from what is acceptable especially in
sexual behaviour, while if the stress is on the second syllable, perˈvert, it turns into verb
that has meaning of corrupt morally or by intemperance or sensuality. Thus, it is
important to know when to stress certain syllable.
The guidelines on word stress combining the propositions of Collin and Mees
(2013: 132) and Giegerich (1992: 184) are: (1) for words consisting of two or three
syllables, the primary stress is on the first syllable but we have to keep in mind that
usually the verbs and adjectives are in final stress and (2) for words consisting of longer
words (four or more syllables), there is a tendency for the antepenultimate syllable (third
from last syllable) to have primary stress.

B.2 Rhythm
All speech is said to have rhythm if it is delivered fluently and without interruption
(Giegerich, 1992: 258). English language is considered to be stress-timed language since
the interval between stressed syllables being isochronous or the interval roughly equal in
time. Sentence stress or the stress of word in connected speech is the basis of rhythm
(Collin and Mees 2013: 136). In speech, the stress will usually fall on the syllable of
content words or lexical words that consist of nouns, main verbs, adjectives, and most
adverbs while function words that consist of articles, auxiliary verbs, verb be,
prepositions, pronouns, and conjunctions left unstressed with few exceptions such as
demonstratives and wh-words. For example, in Giegerich (1992: 259), ˈThis is the ˈhouse
that ˈJack ˈbuilt comprises seven words as well as seven syllables where three words are
content words while the rest are function words. The stress should fall on those three
words (house, Jack, and built) but since this is demonstratives, it should be stressed
which makes it has four stressed syllables. Between the stressed syllables, there is timing
unit known as foot. Time unit to pronounce ˈThis is the, ˈhouse that, ˈJack, and ˈbuilt is
roughly equal due to the nature of English as stress-timed language. Thus, this phrase has
four feet; the first consist of three syllables, the second two, and the last two have one.
This guideline is really useful to analyse English that written in traditional manner (like
rhyming couplets and iambic pentameter) poetry later.

B.3 Intonation
Intonation is the pitch movement of speaker’s voice during an utterance. Unlike
Mandarin that considered being tonal language i.e., the different tone will produce
different meaning, English does not use intonation to differentiate meaning but rather to
use it as a supra-segmental feature.
In any utterance, there will be intonation group where certain syllable that has a marked
change in pitch, which highlights that certain syllable or make the syllable prominence by
uttered it longer and louder. That prominence syllable is known as the nucleus intonation
and it is considered to be an essential component of intonation group. The possible tones
carried by the nucleus are fall (⸌), rises (⸌), fall-rise (˅), and rise-fall (^).
As a supra-segmental feature, intonation has four functions namely the focusing,
attitudinal, grammatical, and discourse (Collins and Mess, 2013). Focusing is a speaker
who focuses on the most significant information by mean of the location of the nucleus.
For e x a m p l e , J a m e s adored his gorgeous new motorbike. The m o s t
significant information here is the new motorbike thus, it uttered longer and louder.
Attitudinal is about a speaker who imposes an attitude on top of the bare semantic
content of what
being said. For a fall-rise (˅) tone, it implies doubt, correction, reservation, and appealing
to the listener to consider while for a rise-fall (^) tone, it implies impressed, arrogant,
confident, self-satisfied, mocking, and putting down. Grammatical function is about
speaker distinguishing certain syntactic relationship for example statement vs question
like You’re are going to ⸌Malaysia. vs You’re are going to ⸌Malaysia?. Discourse is
about the organisation of conversation between two or more speakers. The nucleus tone
in the discourse can be divided into two the falling tones that suggest finality and
unloading of information and rising tones that indicate non-finality and information are
sought or anticipated, rather than unloaded.

METHODOLOGY

This study is a qualitative study on a poem with the emphasis on the supra-
segmental features of the poem. The main data used in this study is Robert Frost’s poem
title Fire and Ice, retrieved from Poetry Foundation’s website. The tools used in the study
were segmental and supra-segmental features. The units of analysis are each word and
line found in the poem. Those segmental and supra-segmental aspects are identified in
each line.

DISCUSSION

The poem was written by Robert Frost (1874 - 1963) and considered to be
interesting since the poem composed in traditional and modern manner. The poem
consists of nine lines varies between two-meter lengths (either eight syllables or four
syllables) and uses three sets of interwoven rhymes, based on -ire, -ice, and –ate. Aside
from the very clear use of rhyme, the poem seems rather formless. The first two lines are
like a prologue. Those lines do not have the same rhythm or syllable count as the other
lines; the first line consists of eight syllables and second line 4 syllables. Starting in line 3
until 7, a pattern becomes clear; eight-syllable lines in an iambic meter, which is
unaccented or unstressed syllable followed by accented or stressed syllable (Wolosky,
2001: 41). It seems that the last two lines are a single line that has been broken in two for
the sake of the rhymes.

A. Segmental Analysis
The analysis started with minimal pairs to tackle the segmental features in the
poem. The minimal pairs found in the poem is and [ænd] in line 1 and end [end] in line 9
which can be distinguished using phonemes /æ/ and /e/ respectively. In weak forms, and
will pronounced as [ənd] or [ən]. Thus, the schwa /ə/ used to designate vowels that have
reduced vowel quality during the realisation of and [ænd] into [ənd] or [ən] in connected
speech. It is important for students to be exposed to the citation forms for a single word
and in a connected speech in order for them to acquire good spoken English language.
As stated previously regarding the difficulties for Indonesians in perceiving the
fricatives, this poem found several words that have fricatives namely the fire (line 1 and
4), I’ve (line 3), desire (line 3), those (line 4), favor (line 4), if (line 5), perish (line 5) and
suffice (line 9). Fire [ˈfaɪ‿ə] (RP) or [ˈfaɪ‿ər] (GA), I’ve [aɪv], favor (or favour in RP)
[ˈfeɪv.ə] (RP) or [ˈfeɪv.ər] (GA), if [ɪf] and suffice [sə.ˈfaɪs] are using phonemes /f/ and /v/
where usually Indonesians will perceive the phonemes as /p/. Desire [di.ˈzaɪ‿ə] (RP) or
[di.ˈzaɪ‿ər] (GA), those [ðəʊz] (RP) or [ðoʊz] (GA), perish [ˈper.ɪʃ], and suffice [sə.ˈfaɪs]
are using phonemes /z/, /ʃ/ and /s/ where usually Indonesians pronounce those phonemes
as /s/. The confusions make sense because the fricative consonants are obstruent, which is
a natural class of sounds that have common articulatory (formed by air passing through a
narrow opening between two parts of the mouth) and phonological feature (Ladefoged,
1993: 62). If we look at the phonetic transcription between RP and GA, the most
prominent difference is the pronunciation of /r/ that usually pronounced in GA while in
not in RP. This makes the GA as rhotic while RP is non-rhotic. For certain word like if, in
RP, this word has no separate weak form but in GA, it may have a weak form [əf]. As we
noticed here, schwa /ə/ is used to unstressed the vowel phonemes of /i/ (tense vowel)
during the realisation in connected speech.
In the poem, there are two digraphs (two letters used together to represent a single
speech sound) words namely the wh- in who (line 4) and -gh- in enough (line 5) that can
be very problematic to pronounce especially for Indonesians. Who is a digraph wh, which
the digraph pronounced as /w/ in most cases (Longman Pronunciation
Dictionary).
However, through the changes in English language, the words like who, whole, and whom
are having reduction, thus, pronounced as huː, hoʊl, and huːm, respectively. Who can be
in strong (huː) and weak forms (hu or u). The weak forms are used only for the relative,
not the interrogative.
Enough is a word that includes in -gh- pattern words. The -gh- pattern words come
from Anglo-Saxon word, for example, tiht (tight). The /h/ was a hard-low-pitched sound.
Then around the 13th century, the /h/ became /gh/ because of the Franco-Latin influence
to try demonstrating this low-pitched sound. By late Middle English period (1150 – 1476),
the /gh/ sound was either dropped or became /f/ sound in northern dialect (Upward &
Davidson, 2011: 181), for example enough pronounced as [ə.ˈnʌf] (RP) or [i.ˈnʌf] (GA),
where /gh/ dropped and become /f/. There is no fixed rule on pronouncing the digraph -
gh- but there are several possible pronunciations like /g/ in ghost [ɡəʊst] (RP) or [ɡoʊst]
(GA), /f/ in enough, and silent after i and other vowel letters as in high [haɪ] and daughter
[ˈdɔːt.ə] (RP) or [ˈdɔːt.ər] (GA) (Long Pronunciation Dictionary).
With the help of pronunciation dictionary like Longman Pronunciation Dictionary,
it is really helpful for the students whom seriously interested in English pronunciation,
specially to study the segmental aspects. In fact, Ladefoged (1993: 76) also suggested
for students to use Longman Pronunciation Dictionary since this dictionary is the leading
authority on contemporary English pronunciation that covers British, American, and other
variants of the world-wide language.

B. Supra-segmental Analysis
The basic for supra-segmental is the consonants and vowels as the single sound or
segment of which speech is composed because supra-segmental is the units of linguistic
which operate above single sounds. Then, the segments will form the syllables that
eventually turn into utterances or connected speech (Ladefoged, 1993: 14). The analysis
on supra-segmental aspects of this poem will be started with the stress that divided into
word stress and sentence stress in each line of the poem. Mostly the words contained in
the poem are monosyllabic word except tasted (line 4), desire (line 4), favor (line 5),
perish (line 5), enough (line 6), destruction (line 7), and suffice (line 9). Of all those
words, only enough (adverb) and destruction (noun) are not verb while the rest are.
Enough [ə.ˈnʌf] (RP) or [i.ˈnʌf] (GA) is adverb which means, the syllable that should be
stressed (marked with “ˈ” symbol) is on the final syllable, similar to other words that
categorised into verbs. For destruction [di.ˈstrʌk.ʃən], the stress will fall on the
penultimate (second from last) position. In sentence stress or connected speech, the
content words which carry a high information load are normally stressed, specifically the
syllable, with few exceptions. Line (1) has 8 words as well as 8 syllables since all of the
words are monosyllabic. There are 5 content words which mean there are 5 stressed
words namely say, world, will, end, and fire. For the function words like some [səm], the
[ðə], in [ɪn] (RP) or [§ən] (GA), the words are unstressed and pronounced in weak forms.
In RP, the word in has no weak form but in GA, there is (Longman Pronunciation
Dictionary). We also notice that in the realisation of the unstressed vowel in connected
speech, schwa /ə/ is used to designate vowels that have reduced vowel quality. Even
though the word those (line 4) is function word, it is stressed in connected speech because
it is a demonstrative word, which is the exception as being stated before. The same
method can be used to analyse the following lines of the poem.
Sentence stress is the basis of rhythm in English. In line (1), there is a total of six
content words as equal to six syllables which make line has six feet. The time interval
between some|, say the|, world|, will|, end in|, and fire| should be roughly equal. The sign
“|” used as a marker of the foot. The rule also applied to other lines. This rule also can be
applied in real life conversation. Lines (3) until line (7) are interesting because they show
an obvious iambic pattern; eight syllables with four accented or stressed syllables. Line
(8) and (9) also show the iambic pattern with lesser syllables. Thus, to recite all those
lines, we can use a different rule with the previous since the lines have iambic patterns.
The pattern would be an unstressed-stressed pattern (Wolosky, 2001: 139).
When reciting the poem, just like producing normal utterances, the pitch changes to
emphasise certain words called the nucleus which carries the most significant information
in a tone group. In line (1), there are two nucleuses found which are world and fire, which
make this line has two tone groups. Thus, while reciting line (1), there will be great
change in pitch when pronouncing the words world [wɜːld] (RP) or [wɝːld] (GA) and fire
[ˈfaɪ‿ə] (RP) or [ˈfaɪ‿ər] (GA). The pitch changes (fall pitch) that begin from world
continue only until the beginning of the next tone group, namely the fire. This in only the
guideline since there is no exact rule on defining which one will be the nucleus; all
depends on the speaker emotional state, and to some extent, personal psychological
characteristics (Ladefoged, 1993: 257).

CONCLUSION

In summary, the poem consists of both segmental and supra-segmental features. To

recite the poem, the knowledge of segmental (consonant and vowel phonemes) and supra-
segmental features (stress, rhythm, and intonation) is needed.
The implications for this study are students or readers will familiar with the
segmental and supra-segmental features found in a poem and know how to recite any
poem later. This is important as part of the learning process to acquire good spoken
English.

REFERENCES

Alip, F. B. (2016). Language: Theory, acquisition, and application. Yogyakarta: English

Letter Department Universitas Sanata Dharma.
Collins, M., & Mees, I. M. (2013). Practical phonetics and phonology (3rd ed.). New
York: Oxford University Press.
Frost, R. (1920). Fire and Ice. Retrieved June 27, 2017, from
https://www.poetryfoundation.org/poems-and-poets/poems/detail/44263
Giegerich, H. J. (1992). English phonology: an introduction. Cambridge: Cambridge
University Press.
Ladefoged, P. (1993). A course in phonetics (3rd ed.). USA: Harcourt Brace Jovanovich.
Poole, S. C. (1999). An introduction to linguistics. New York: St. Martin’s Press.
Rudling, J. (n.d.). Words with the -gh- letter pattern. Retrieved July 8, 2017, from
https://howtospell.co.uk/gh-words
Shmoop Editorial Team. (2008). Fire and Ice Form and Meter. Retrieved June 27, 2017,
from http://www.shmoop.com/fire-and-ice-frost/rhyme-form-meter.html
Upward, C., & Davidson, G. (2011). The history of English spelling. UK: Wiley –
Blackwell.
Wolosky, S. (2001). The art of poetry: How to read a poem. New York: Oxford
University Press.

APPENDIX

Fire and Ice by Robert Frost

(1) Some say the world will end in fire,
(2) Some say in ice.
(3) From what I've tasted of desire
(4) I hold with those who favor fire.
(5) But if it had to perish twice,
(6) I think I know enough of hate
(7) To say that for destruction ice
(8) Is also great
(9) And would suffic