Why Your Mandarin Program Isn't Working (And What James Wrenn Knew About That Decades Ago)
I've been around university Chinese programs long enough to watch the same cycles repeat. We bought into communicative approaches because they sounded modern, then complained students couldn't read a text message from their grandmother. Then we pivoted to input-based instruction, and suddenly everyone's reading graded readers but can't order coffee in Taoyuan. I found myself going back to older materials more often than I cared to admit, and that's how I ended up revisiting Chinese Language Teaching In The United States James J Wrenn's work, which keeps getting overlooked in favor of whatever method is trending in the ACTFL guidelines. Wrenn's approach wasn't particularly revolutionary on paper. He argued for a balanced treatment of skills, which sounds obvious until you realize most curricula spend roughly 70 percent of instructional time on speaking and listening while treating reading and writing as extracurricular activities students are expected to pick up on their own. His point was structural: Chinese as a language demands explicit attention to the writing system from day one if students are ever going to be functionally literate, and you can't realistically separate the four skills because Chinese characters don't have a phonological transparency that lets you sound things out the way you can with alphabetic languages. I remember working with a second-semester student who could hold a decent conversation about her weekend plans but couldn't identify the characters for those same activities when she saw them written down. This is not a hypothetical edge case. This is the default outcome when programs follow the standard model. Students hear "zou" and know it means walk, but show them and they have no idea what they're looking at. Wrenn called this the two-Chinese problem, and honestly, it hasn't gotten much better since he wrote about it.
What Actually Works When You Implement This Approach
The practical upshot of Wrenn's framework is that you introduce characters alongside pinyin and spoken forms from the beginning, but you don't treat them as separate tracks. I used to make the mistake of doing pinyin isolation drills before character recognition, thinking students needed phonological grounding first. That added maybe three weeks to the curriculum with no measurable improvement in reading outcomes. Once I started teaching the character components and the phonetic elements within them simultaneously with the oral forms, students who'd been struggling with tone recall started improving because they had a visual anchor to attach the sound to. The counter-intuitive part that nobody mentions is that heavy character work early on actually improves listening comprehension. When students can map spoken syllables to written forms quickly, their auditory processing becomes more efficient. They stop sounding out every word in their head and start recognizing whole compounds. I ran this past a colleague who teaches cognitive linguistics and he said it lined up with what they know about orthographic depth affecting phonological awareness. The data in the Chinese teaching literature on this is thinner than you'd think, but the classroom observation is consistent enough that I stopped questioning it.
The Hard Parts No One Warns You About
Implementing a balanced four-skills approach in an actual American university setting runs into structural walls. Most programs operate on a twelve-week semester with meetings three times a week. That gives you roughly forty-eight contact hours per level. Wrenn's model assumes more time on task than most curricula can deliver without cutting something else. I found that dropping pop quizzes and restructuring homework assignments to require less grading overhead freed up about six hours per level for intensive character writing practice, which made the difference between students achieving basic literacy and them remaining dependent on pinyin annotations. There's also the assessment problem. Standard placement tests and proficiency exams in the US still weight speaking heavily. A student who reads at a high-intermediate level but has a timid speaking voice might test out of your program entirely, and then you've lost them. I started building my own reading and writing checkpoints that feed into the final grade separately from oral performance, which helped identify students who were being under-placed because of accent anxiety rather than actual ability. It took some political capital with the department chair to get that accepted, but once it was in place, retention improved noticeably.
Get the Full Details
A Real Workaround I Had to Figure Out
Here's a specific problem that nearly broke my implementation of this approach: advanced students who'd been in the program for two years still couldn't navigate a simple Chinese newspaper without reverting to pinyin or asking for translations. They'd spent so much time on speaking fluency that their character recognition was at a functional minimum. The workaround was introducing a weekly "newspaper lab" where students worked in pairs to decode a short article without using any technology. No phones, no translators, just pencil and paper. The first few sessions were brutal. Students were genuinely distressed. But after about eight weeks of this, the speed and accuracy of their independent reading improved by roughly forty percent based on my timing logs. I also recommend having students keep a character notebook where they record new compounds they encounter in reading, with the semantic and phonetic components identified. It's old school, it takes time, and students will complain about it initially. The tradeoff is real. But I've seen multiple cohorts where students who maintained these notebooks consistently through the year were the ones who made it to intermediate-high in reading and writing on the ACTFL scale, while the ones who didn't maintain them stalled out at low-intermediate regardless of their speaking ability.
When This Approach Fails and What to Do Instead
Wrenn's model doesn't work well in contexts where students have very limited contact hours with the language. If your program is meeting once a week for a full year, or if students are in a heritage learner track with strong speaking but zero literacy, forcing a balanced four-skills approach from week one will frustrate everyone. In those cases, I've had better luck with a modified version that prioritizes literacy for heritage speakers and prioritizes communicative competence for complete beginners, then slowly introduces the missing skill area as the semester progresses. It's not as elegant as Wrenn's original framework, but elegance doesn't matter when you're working with twenty-minute-per-week students who show up late to class every Tuesday. The larger institutional problem is faculty capacity. Teaching a balanced approach requires instructors who are comfortable with all four skills, and finding or training those people is harder than it sounds. Many program directors assign teachers to levels based on their specialty, which recreates the skill imbalance at the institutional level regardless of what any pedagogical framework recommends. I've seen programs try to solve this with team teaching, and it works when both instructors are on board, but coordinating schedules and lesson plans between two faculty members adds administrative overhead that most departments are already too thin to absorb. If you're looking at Wrenn's work for implementation advice, the practical takeaway is simpler than the theory suggests: stop treating characters as optional. Start from day one with a system that connects sound, form, and meaning together. Grade reading and writing separately from speaking so students who are strong in different areas aren't penalized. And accept that you'll need to make room in the schedule for character work, which means cutting something else, usually something that doesn't add as much long-term value to student outcomes.
