A university-ready program design that treats time-on-task, cognitive load, and proficiency measurement as first-class constraints—not afterthoughts.

The non-negotiable premise

If the institution’s goal is high functional proficiency, the program must be engineered around exposure hours + spaced retrieval + speaking density, not around textbook coverage. Category-IV-hour scales (thousands of hours for “professional” proficiency) make the gap visible: a standard 4-year sequence cannot rely on classroom time alone. (state.gov)

1) Outcomes and targets (what success means)

Proficiency targets by track (realistic but ambitious)

Use ACTFL OPI/OPIc + AAPPL (speaking/listening/reading) as primary, with HSK as secondary (useful but not the north star).

  • End of Year 1: Novice High → Intermediate Low (speaking/listening), basic literacy foundation
  • End of Year 2: Intermediate Mid (speaking), Intermediate High (listening/reading)
  • End of Year 3: Intermediate High / Advanced Low (speaking), Advanced Low (listening/reading)
  • End of Year 4: Advanced Low → Advanced Mid (speaking) for top half; Advanced Low+ (listening/reading) as standard

This aligns with observed “reading advances faster than speaking” patterns and makes speaking an explicit program outcome rather than a hoped-for byproduct. (l2trec.utah.edu)

2) The hour model (how you close the gap)

Minimum weekly “dose” (per student)

Class time alone is insufficient for advanced speaking. The program must require and verify out-of-class input and interaction.

A workable target (non-immersion semesters):

  • Year 1–2: 8–10 hrs/week total Chinese time (class + required input + retrieval + speaking)
  • Year 3–4: 12–15 hrs/week total
  • Plus one intensive summer/immersion block strongly recommended (or required for “Advanced track”)

Why this isn’t optional: authentic media engagement is repeatedly linked with stronger outcomes; programs that don’t structurally build it are leaving proficiency to chance. (l2trec.utah.edu)

3) Curriculum architecture (what gets built)

Three parallel “systems,” with explicit load management

Chinese requires simultaneously building:

  1. Sound system (tones, syllables, perception/production)
  2. Meaning system (lexicon + grammar patterns + discourse routines)
  3. Script/literacy system (characters, reading fluency, typing, writing)

The failure mode is letting script demands cannibalize input and speaking time—especially early—when tone acquisition is fragile under cognitive load. (researchgate.net)

4) Sequencing plan (what happens when)

Year 1: “Sound-first + input-first + typed literacy”

Goal: prevent tone/phonology from becoming permanent debt.

A. Tone/perception training (daily micro-doses)

  • 10–15 minutes/day structured tone perception + minimal pair discrimination
  • 10 minutes/day guided production (shadowing + feedback)
  • Weekly short “tone check” (low-stakes, frequent)

Rationale: tone perception is sensitive to cognitive load; separate it early from heavy script burden. (researchgate.net)

B. Input quotas (required, not suggested)

  • Listening: 90–120 min/week graded audio/video
  • Reading: 45–60 min/week graded readers (start small, scale fast)
  • Require proof: short comprehension checks + discussion prompts (not journaling)

C. Script policy (stop pretending handwriting is the gateway)

  • Teach typing + recognition as primary literacy skills in Year 1
  • Handwriting limited to:
    • radicals/components drills for structure awareness
    • a small “core” set for identity + confidence (e.g., 150–250 chars)
  • Move “beautiful handwriting” to optional electives

Rationale: writing-system structure matters, but heavy copying early can be an efficiency trap; literacy should scale. (files.eric.ed.gov)

D. Orthography sequencing (adaptive, not ideological)

Use pinyin strategically early, then phase it down:

  • Weeks 1–6: pinyin heavy for phonology + rapid vocabulary uptake
  • Weeks 6–12: mixed pinyin + characters
  • Semester 2: character-forward for reading, but keep pinyin as scaffold for new words where needed

Rationale: orthography affects tone/perception/learning; the point is cognitive load management, not purity. (ling.cuhk.edu.hk)

Year 2: “Reading fluency + speaking density”

Goal: keep speaking from lagging behind literacy.

A. Speaking density requirement

  • Add mandatory conversation lab (small groups) weekly
  • Require minimum speaking minutes per week per student (tracked via structured tasks)

B. Literacy scaling via frequency + structure

  • Characters taught in frequency bands, but with explicit component/radical structure instruction
  • Prioritize recognition → reading speed → comprehension, not copying volume

C. Retrieval architecture (make memory unavoidable)

Language learning collapses without spaced retrieval.

  • cumulative quizzes weekly
  • “spiral” vocabulary: old items reappear by schedule, not by chance
    Spacing/retrieval effects are robust in learning science; universities should institutionalize them. (augmentingcognition.com)

Year 3: “Domain competence + long-form input”

Goal: build sustained comprehension and discourse ability.

A. Extensive reading becomes core

  • 120–180 min/week reading (graded → semi-authentic)
  • reading is tied to:
    • short oral summaries
    • argument/response tasks

Vocabulary coverage matters for comprehension; programs must force the volume required to reach it. (lextutor.ca)

B. Content-based modules (Chinese through something)

Offer 2–3 “content lanes,” e.g.:

  • Chinese media & society
  • business & negotiation
  • science/tech discourse
  • classical-to-modern bridge (optional)

C. Tone and pronunciation “maintenance”

Tone work doesn’t end; it shifts to:

  • targeted repair
  • fluency under speed
  • listening under noise

Year 4: “Advanced speaking pathway + capstone verification”

Goal: ensure advanced speaking is not a rare accident.

A. Two tracks in Year 4

  • Standard track: Advanced Low in speaking target
  • Advanced track: Advanced Mid target + immersion/intensive requirement

B. Public-facing capstone

  • debate/presentation + Q&A
  • recorded portfolio
  • external rater or inter-institutional moderation

This directly counters “narrative governance” (grades) with verifiable proficiency artifacts.

5) The heritage/non-heritage split (mandatory redesign)

Mixed classrooms predictably mis-serve both groups because skill profiles differ (often stronger oral skills for heritage, uneven literacy). You need either:

  • separate tracks, or
  • a shared class + separate labs and assessments

Heritage divergence is documented and shouldn’t be treated as a “minor accommodation.” (brill.com)

6) Assessment system (what you measure, you get)

Required measurement points

  • Entry placement (separate heritage placement)
  • End of each year: ACTFL-aligned speaking/listening/reading
  • Mid-semester quick checks: tone + listening + fluency

What you stop doing

  • Stop letting “chapter coverage” function as the real assessment.
  • Stop letting handwriting volume function as a proxy for literacy.

7) Institutional reforms (the uncomfortable but necessary ones)

A. Staffing and section design

If you want advanced speaking, you must buy speaking time:

  • smaller speaking labs
  • more TA conversation sections
  • structured peer conversation systems with accountability

B. Make input visible in the transcript (seriously)

Create “Input Lab” credits or requirements. If students must do 2–4 hours/week of media/reading to succeed, it should be institutionalized like science labs.

C. Build a “CSL Program Dashboard”

Publish annually:

  • proficiency distributions by year
  • % reaching targets by track
  • retention/drop-off
  • effect of immersion participation

This is how you prevent self-deception and drift.

8) What this blueprint is optimized to fix

Confirmed failure patterns it directly attacks: