TL;DR
Vietnamese learners of German struggle to distinguish the German consonant cluster /ft/ from single consonants, particularly in coda position. This difficulty is attributed to the limited auditory cues available for the plosive in that position, and surprisingly, the learners' proficiency in English as a second language seems to aid their perception of other German contrasts.
What Problem Does This Paper Address?
Many studies have noted that Vietnamese speakers learning German as a foreign language (GFL) have difficulties producing German consonantal clusters. This paper investigates whether this problem extends to the perceptual level, specifically asking if Vietnamese learners can even hear the difference between a German consonant cluster and its constituent single consonants. The researchers focus on the German cluster /ft/ and its contrast with /f/ and /t/, examining perception in both word-initial (onset) and word-final (coda) positions. They hypothesize that coda position will be more challenging due to Vietnamese phonotactic restrictions.
Methods
The study employed an AX discrimination task using the software Praat. Participants were presented with pairs of real German words and asked to determine if the words were identical or different. The stimuli included minimal pairs of real words (nouns, verbs, prepositions) that contrasted:
- The cluster /ft/ with either /f/ or /t/ in both onset and coda positions.
- Single consonants in onset and coda positions.
All words were monosyllabic to minimize memory load. The audio stimuli were produced by a native German speaker. Participants' responses were analyzed using generalized linear mixed-effects models to assess the impact of contrast type (cluster vs. single consonant), position (onset vs. coda), and the number of months of German tuition.
Datasets
The study used a custom-created dataset of 215 stimulus pairs, comprising:
- 36 minimal pairs of real words contrasting the /ft/ cluster with single consonants (/f/ or /t/) in onset or coda position.
- 38 minimal pairs of monosyllabic words without clusters, contrasting single consonants in onset or coda position.
These stimuli were recorded by a native German speaker. No pre-existing datasets were used for training or evaluation.
Key Insights and Ideas
- Coda position is the real challenge: The most significant difficulty for Vietnamese learners was distinguishing the /ft/ cluster from the single consonant /f/ when they appeared at the end of a word (coda position).
- English L2 as a facilitator: Surprisingly, learners performed well on most other contrasts, including /ft/ vs. /t/ in coda position. The authors suggest that their proficiency in English, which allows consonant clusters in both onset and coda positions, may have transferred and helped them acquire German phonological contrasts as a third language (L3).
- Auditory cues matter: The difficulty with /ft/ vs. /f/ in coda position is linked to the weak auditory cues available for the plosive component of the cluster in that position. The friction noise that distinguishes /ft/ from /t/ is more salient.
- No effect of tuition duration: The number of months learners had studied German did not significantly impact their perceptual performance, suggesting that initial exposure is crucial, but further study didn't necessarily improve these specific perceptual distinctions.
Key Results
- Participants performed significantly better in onset position than in coda position for distinguishing the /ft/ cluster from single consonants.
- The contrast between /ft/ and /f/ in coda position was significantly more difficult than other contrasts, including /ft/ vs. /t/ in coda position and all onset contrasts.
- Overall accuracy for all stimulus pairs was 87.34% (s.d. 10%).
- There was no significant difference in performance between comparing the cluster to a fricative (/f/) or a plosive (/t/) when the contrast was in onset position. However, in coda position, the /ft/-/f/ contrast was problematic.
Limitations and Future Work
- Grammatical contrast ambiguity: The authors acknowledge that some word pairs used to test the /ft/-/f/ coda contrast were identical verbs differing only in grammatical inflection (indicative vs. imperative). They suggest that learners, lacking sufficient experience with German grammar, might have perceived these as the same lexical item, potentially inflating error rates.
- Production data needed: The paper focuses solely on perception. Future work will involve analyzing the production of these stimuli by the participants to gain a more complete understanding of the challenges.
- Further investigation of L2 influence: While English L2 influence is proposed as a reason for good performance on other contrasts, more research could explore this cross-linguistic transfer in detail.
- Listener ratings: The authors plan to conduct listener ratings on the production data to triangulate findings and further understand the perceptual difficulties.
Summary updated