TL;DR
This study uses corpus linguistics to compare how Chinese college students and native English speakers use "lexical chunks"—prefabricated phrases like "as far as I’m concerned"—during public speaking. While both groups use similar broad categories of phraseology, Chinese students rely on a much narrower range of expressions and often use longer, more formal phrases to "buy time" or sound more sophisticated, sometimes missing the subtle cultural nuances that native speakers navigate intuitively.
What Problem Does This Paper Address?
In the high-pressure environment of impromptu public speaking, speakers don't just build sentences word-by-word; they rely on "phraseology"—multi-word units that function as single building blocks. For English as a Foreign Language (EFL) learners, mastering these chunks is the key to fluency and "naturalness."
However, most research has focused on how EFL students write. This paper addresses the gap in oral public speaking. The author investigates whether Chinese college students use the same types of linguistic "shortcuts" as native speakers and identifies where their usage becomes repetitive, unnatural, or culturally disconnected.
Methods
The researcher employed a corpus-driven contrastive analysis.
- Extraction: Using the software AntConc, the author extracted N-grams (sequences of 2 to 6 words) that appeared with high frequency.
- Filtering: Sequences that lacked complete meaning (e.g., "and I believe the") were manually removed.
- Classification: The remaining chunks were categorized using the Nattinger & DeCarrico (1992) framework, which divides phraseology into four structural types:
- Polywords: Short, fixed phrases (e.g., "by the way").
- Phrasal constraints: Phrases with "slots" (e.g., "a [year] ago").
- Institutionalized expressions: Sentence-length fixed units (e.g., "thank you").
- Sentence builders: Frameworks for entire ideas (e.g., "I think that X").
Datasets
The study compared two primary datasets:
- CLEPSC (Chinese Learner English Public Speaking Corpus): A self-built corpus of 300 impromptu speeches delivered by contestants in major Chinese English speaking competitions (2010–2020), totaling approximately 130,800 words.
- Reference Corpus: 55 speeches by native speakers selected from the American Rhetoric (AR) corpus, carefully curated to match the size of the CLEPSC (~131,000 words) to ensure a fair statistical comparison.
Key Insights and Ideas
- The "Advanced" Trap: Chinese students often use long, complex chunks like "as far as I am concerned" because they perceive them as "advanced" or idiomatic. Native speakers, following the Economy Principle, tend to replace these with shorter, more efficient phrases like "in my view" or "in my book."
- Chunks as "Time-Buyers": The study highlights that for EFL learners, long chunks aren't just for meaning—they are cognitive crutches. They allow the speaker to keep talking while their brain plans the next complex sentence, which is crucial in impromptu settings.
- Socio-Pragmatic Blind Spots: There is a notable difference in how the two groups evaluate ideas. Chinese students frequently use "It’s a good idea to..." while native speakers prefer "I’m a great believer in..." The author notes that the latter carries religious/cultural connotations (the word "believer") that Chinese students might avoid or be unaware of, leading to a more "textbook" and less "natural" sound.
- Lack of Diversity: While the quantity of chunks used by Chinese students is high, the variety (type) is low. They over-rely on a few "safe" patterns, such as using "my topic today is..." significantly more often than native speakers, who use a wider array of topic-introducing phrases.
Key Results
- Structural Similarity: Both groups showed a strong preference for sentence builders, which accounted for roughly 50% of all chunks used.
- Frequency Disparities: Chinese students used the closing chunk "thank you" at a much higher ratio (accounting for 6.61% of all their chunks) compared to native speakers (0.92%).
- Topic Markers: The specific phrase "my topic (today) is (that) X" appeared 33 times in the student corpus but only 5 times in the native speaker corpus.
- Polyword Errors: Students were found to make more errors in fixed phrases, such as saying "in the campus" (11 instances) instead of the idiomatic "on the campus."
Limitations and Future Work
The author acknowledges two main limitations:
- Corpus Size: At ~130k words, the corpus is relatively small for linguistic generalizations.
- Proficiency Bias: The CLEPSC consists of speeches from high-level competition winners. Therefore, the findings might not represent the "average" Chinese college student, whose English proficiency may be lower.
Future Work: The author suggests expanding the research to students with lower English proficiency to see if the same patterns of over-reliance and lack of diversity hold true across different skill levels. There is also a call for pedagogical shifts—moving from teaching fixed "routines" to teaching "patterns with open slots" to help students gain flexibility.
Summary updated