General Linguistics

Assignment 1
Ma’am Sadia Asif
Submitted by: Nimra Saeed
Reg. no: F2024084003
M.Phil. (Applied Linguistics)
Semester 1
Summary of the research review of Amreen and To 2020.
Ambreen and To (2020) have conducted a review article on phonological development in
Urdu-speaking children. The study aims to find the researched works and their findings
related to the topic under study. The studies from January 1980 to March 2020 were taken
into consideration, and five methods of the preferred reporting items for systematic review
and meta-analysis were applied. A descriptive review of the available research is provided.
McLeod and Crowe (2018) led a cross-linguistic review to find the pattern of consonant
acquisition in young speakers of 27 languages. They discovered that phoneme interaction
causes phonological development variation and varies from speaker to speaker of various
languages. This demands the need to develop a construct to aid children in phonological
development (as cited in Ambreen and To 2020).
Ambreen and To (2020) explain that Urdu has only two tools for the study of phonological
development and those tools are not widely accessed because of their limitations. This urges
the competent personnel to create a tool that can help in the speech therapy of young Urdu
speakers in an unbiased manner. Speech-language pathology service in Pakistan is not an old
practice, but started in Karachi in 1983 [(Shafqat,2020) as cited in Ambreen and To, 2020]
this was a starting point that led to the opening of further clinics and on the national level,
higher degree programs in this field were initiated. Speaking of the Urdu phonological
system, the reviewers write that it is the lingua franca of multiple communities and is
mutually intelligible with Hindi because of the pre-partition scenario.
Urdu has 43 consonants which vary from one another due to their manner of articulation
[(A.M. Saleem et al. (2003) as cited in Ambreen and To 2020]. It also has 23 vowels and 15
diphthongs (Centre for Language Engineering, 2015). The reviewers mentioned that this
study finds vowel and consonant acquisition as well as phonological errors in speakers of
Urdu aged till 8 years. In phase one of the research, reviewers identified the relevant
keywords from the literature, and phase two was for screening of relevance. In phase three,
the studies that matched their criteria of research elements were included. The inclusion
criteria are given on Page 3 of this review article. Data was extracted from a form that was
originally developed by a Cochrane collaboration (version 3; Cochrane Collaboration,
2014a). For Quality appraisal, the CAT-CSS (Soliman et al., 2017) was used and studies were
quantitatively graded on a scale that ranged from poor to excellent. This degree was assigned
based on their quality. All the chosen studies were cross-sectional except one which was
synchronic. The inclusion and exclusion criteria were not reported in some of the studies.
Similarly, important demographic information was also missing from all of the published
studies on Urdu phonological development.
Two studies that are reviewed in this analysis were focused on typically developing children
and the other targeted a group of children with articulation and phonological disorders
(Noveen, Butt, & Alam, 2017; Noveen, Masood, & Habibullah, 2017). The rest of the studies
were aimed to study children with no evident disorder of speech or language. Children with
articulatory or phonological disorders were not included in the unpublished studies.
Moreover, the published studies did not include the demographic characteristics of the
participants. However, this information was there in unpublished theses. The sample size of
these studies ranged from 10-120 and the selection of the sample was based on convenience.
The number of words for data collection in these studies ranged from 60 to 95. All the
participants were located in different areas of Pakistan. The mentioned studies were
conducted in the last decade. This can be due to the substantial growth of this discipline
during that period.
Sharif (2015) was the only researcher to use two different kinds of word samples for the two
groups of children. The younger ones were given monosyllabic words and the older ones
were provided with polysyllabic words. However, the number of words was same for the both
groups; 65. All of the studies recorded the responses of participants in the form of audio or
video which were later transcribed with the IPA method for research. The process of data
collection revealed that all the studies were majorly concerned with phonological errors and
development along with consonants. Vowels and diphthongs have not been studied at all in
the research.
One of the major findings of the studies under review was sex differences. As cited in Amber
and To (2020), the studies of (Abbasi, 2017; Ali, 2017; Mumtaz, 2017; Shabbir, 2016; Sharif,
2015; Zahra, 2016) shed light on this area of development. Only, Sharif (2015) was the one to
report the phonemic acquisition differences of the both sexes in the separate tables. It was
found that boys and girls showed similar patterns in constant acquisition as well as
phonological processes (Mumtaz, 2017), 3;01–4;12 (Zahra, 2016), and 5;01–6;12 (Ali, 2017).
However, the girls were found more inclined toward anaptyxis (insertion of a vowel between
a consonant cluster) rather than boys (Abbasi, 2017; Shabbir, 2016).
Another major finding was speech sound acquisition which increased as the age band of
children under study increased. Children were divided into age groups and the expertise in
phoneme acquisition and production was studied. The criteria were based on correct sound
production and the percentage of children who correctly pronounced the given words (Zahra,
2016) as cited in Ambreen and To (2020).
Phonological processes were also a major finding in this review. The study of Noveen et al.
(Noveen, Butt, & Alam, 2017; Noveen, Masood, & Habibullah, 2017) talked about
substitution, omission, addition as well as distortion of phonemes. Vowel substitution was the
most recurrent process. All children aged between 3 to 5 used vowel substitution.
Assimilation processes varied in children and among the ages of three to four nasal
assimilation was most prevalent. The deletion of the final consonant was also noticed
repeatedly by the children of 3-4 years age group. The aspirated sounds were also frequently
deleted by the participants. Developmental trends were also a focal point of the few
researches. the stops and nasal consonants were acquired much quicker as compared with
other consonants (sharif, 2015) as cited in Ambreen and To (2020). Accuracy patterns were
reported following the PCC and PPC patterns C (Ali, 2017); Mumtaz, 2017; Zahra, 2016).
The summary of research findings by Ambreen and To (2020) is briefed further in major
categories along with the references. The speech sound acquisition was studied and it was
noticed that the Urdu language follows the universal pattern for consonant sound acquisition
(McLeod & Crowe, 2018). Sharif (2015) reported that the voiced consonants of voiced labial
and dental stop categories were acquired quickly as compared to the voiceless consonants.
The early acquisition of phonemes with the easier manner and place of articulation is a
property of several languages. However, except for some language-specific trends, almost all
languages follow the universal acquisition pattern worldwide (Green et al., 2000).
In the Urdu language, aspiration of consonants is also an important phenomenon and these
are acquired later than the unaspirated consonants (Sharif, 2015). The Cantonese language
shows the same pattern for aspirated and unaspirated consonants (Cheung,1990). In the Urdu
language, the retroflex r sound is produced at customary (4 y/o) and mastery (5.5 y/o) levels.
At 7 years of age, which is the oldest age studied for Urdu consonant acquisition, three
consonants, /ɡh, ɣ, ɖh /, had still not been acquired, and 12 consonants were not fully
mastered. This finding proves that the acquisition and mastery of the Urdu phonetic inventory
continue after 7 years of age according to Ali (2017).
Among the three studies that examined phonological processes in Urdu-speaking children,
two (Abbasi, 2017; Shabbir, 2016) employed the same categorization for the processes
studied. The third study (Khan, 1984) addressed the presence or absence of several
phonological processes but did not report their percentages or prevalence. Consequently, it
was not possible to determine which processes were more common between ages 1;08, and
2;06. Shabbir (2016) and Abbasi (2017) used similar broad categorizations and sub-
categorizations, of the phonological processes. This technique made it easier for reviewers to
review their data and determine which processes were more prevalent and which gradually
disappeared across the older age groups.
Giving the final review, Ambreen and To (2020) discussed participants, sample size, age
range, sex differences, elicitation method, transcription system, production criteria,
implications, and limitations. For the inclusion of participants, some studies included
typically developing children while others included children with phonological disorders.
This review comprises of the typically developing children only. These studies aimed to test
the validity of an assessment tool (TAAPU) in both groups of children by comparing them.
The criteria for the inclusion and exclusion of the participants have not been mentioned in
these studies. Furthermore, a sampling frame that considers significant parameters such as
sex and age can also facilitate later data analyses and normative data presentation (Peña et al.,
2006; Plante & Vance, 1994).
The sample size ranged from 10-120. The size of the group directly affects the results and
validity of that study. McCauley and Swisher (1984) and Plante and Vance (1994)
recommended that phonological development studies aiming to report norms have at least
100 participants in each subgroup to obtain reliable data. This can make the data valid and
results sustainable. The interval time between the studies ranged from 2 months to 4 years.
However, Li & To, 2017; McLeod & Crowe, 2018 recommend dividing the study groups at
6-month intervals to detect changes that occur during the acquisition of language. Also, the
limit of 7 years in these studies showed that the phonological development is not complete.
This gap can be searched on in further research to check when this development is completed.
Most studies on phonological development in Urdu-speaking children considered sex
differences. Sharif (2015) specifically examined consonant acquisition by gender, finding that
girls acquired more consonants than boys from 8 to 30 months. The other studies on
consonant acquisition did not make this comparison but analyzed speech sound accuracy,
finding no significant sex differences across age bands from 2 to 7 years. Similarly, studies on
phonological processes noted no significant sex differences in occurrence between ages 3 and
7, except for anaptyxis, as mentioned above. These findings suggest that speech sound
accuracy and phonological processes are largely similar for boys and girls, but future studies
should further explore phoneme acquisition differences to determine if separate norms for
each sex exist.
Only Khan (1984) used spontaneous speech for eliciting language samples, while other
studies relied on structured methods, such as picture naming, which provided a more
controlled, standardized approach. Li and To (2017) advise using a structured tool with a
comprehensive wordlist covering all phonemes and syllable positions which would enhance
sample reliability. Most studies used a wordlist by Shabbir (2016) and Zahra (2016),
including simple-to-complex syllable structures, but often gave participants only one attempt
per word and lacked data on phoneme occurrences. Some studies also did not validate their
tools for content reliability, which may impact result interpretation. Future research should
develop well-balanced, language-specific elicitation tools that consider phoneme occurrences,
and provide clear guidelines on word production attempts. Guidelines by Eisenberg and
Hitchcock (2010) are a great source to ensure consistent phoneme representation in Urdu
phonological studies (as cited is Ambreen and To (2020)).
Ambreen and To (2020) while reviewing the transcription system in the review write that
most studies on Urdu phonological development used IPA for transcription, except for
Mumtaz (2017), who employed CI-SAMPA for easier data entry on computers. For better
transcription efficiency, the use of Phon software (Rose et al., 2006), which supports IPA and
minimizes human error, is recommended. Although some studies double-checked
transcriptions for reliability, none reported inter- or intra-rater reliability. Future research
should ensure interrater reliability by involving independent transcribers and providing
detailed reliability assessment methods. Also, regarding the production criteria Ambreen and
To (2020) state that some studies on Urdu phonological development did not specify criteria
for assigning speech sounds to age groups, while others showed variation in criteria. Most
used 50% for customary and 75% for mastery levels, though Mumtaz (2017) applied 50% for
customary, 90% for mastery, and 75% for acquired levels. Such variability highlights the
need for standardized criteria across studies to enable better comparisons within and across
languages. Additionally, studies should clarify factors like phoneme occurrences and
production opportunities, which may impact findings.
The reviewers have shared the limitations as well as the implications of their work. The Urdu
phonological development research over the past four decades is very limited. This review
assesses the quality of existing evidence and highlights the need for further research to
understand Urdu-speaking children’s phonological development. It collects data on phoneme
acquisition and speech accuracy, aiding clinicians in applying these findings in practice. Both
published and unpublished studies were included, and quality was assessed by two
independent reviewers, though subjective tools and study differences may have influenced
results. Limitations such as the lack of tools for measurement of evaluating factors and
reliability reporting were noted as potential biases in findings.
On a concluding note, the reviewers mention that this work studies the current research on
Urdu phonological development and highlights the need for more studies in this area. Future
research should include children over 7, study vowel development, and use stronger methods,
such as tracking multiple instances of phoneme use for accuracy. Most of these studies were
conducted in the Punjab province of Pakistan. This urges future researchers to look into the
phonological development in children of other areas of Pakistan as well.