Paper and screen both work for Chinese writing practice, and they are good at different halves of the job. Paper gives better feedback about proportion and pressure. A screen can hide the character, check your stroke order and schedule reviews. The productive answer for most learners is both, with a clear division of labour rather than a loyalty.
What paper does better
Proportion feedback. A grid page shows immediately when a character drifts off centre, when the left component is too wide, or when complex characters are being written larger than simple ones. A screen flattens some of that, particularly on a phone.
Pressure and stroke quality. A pen responds to how hard you press and how fast you move, which is part of what makes a stroke look right. Most screens do not reproduce this, and a finger reproduces it least of all.
A visible record. A page of attempts with the failures circled is a physical artefact you can compare against a page from a fortnight ago. That comparison is one of the better motivators available, and a dashboard number is a poor substitute.
No temptation to check. Paper has no reveal button. That sounds trivial and it removes the single most common way practice degrades, which is glancing at the answer before attempting.
What a screen does better
Hiding the character. A notebook cannot stop you seeing the model you wrote three lines ago, and an app can withhold it entirely until you have committed to an attempt.
Checking stroke order. Paper records the finished shape and nothing about the route you took, so a wrong sequence can persist for years unnoticed. An app sees the sequence, which matters because order drives fluency and legibility, and research on how stroke sequences are presented in learning suggests that how this support is given affects what learners retain.
Scheduling. Deciding which of 400 characters to review today is tedious by hand and instant for software, and distributed review is what produces retention rather than volume, as the spaced learning research consistently finds.
Availability. Practice happens in the gaps: a queue, a train, five minutes before a class. A phone is in your pocket and a notebook usually is not.
| Job | Paper | Screen |
|---|---|---|
| Hiding the target | Poor | Excellent |
| Stroke order checking | None | Good |
| Proportion feedback | Excellent | Moderate |
| Pressure and stroke feel | Good | Weak, especially with a finger |
| Scheduling reviews | Manual and tedious | Automatic |
| Practising in spare moments | Rarely to hand | Always available |
| Visible record over time | Excellent | Usually just numbers |
The division of labour that works
Screen daily, paper weekly.
The daily ten minutes runs on a screen: prompt first, character hidden, attempt, stroke order checked, misses scheduled sooner. That covers recall, order and scheduling, which are the parts that decide whether characters stay.
The weekly session runs on paper: a page of the characters you have been practising, written large on a grid, then a few sentences. That covers proportion, stroke feel and writing in sequence, which are the parts a screen serves less well.
This split also solves the motivation problem in both directions. The daily screen session is short enough to happen on bad days, and the weekly paper page is satisfying in a way that a completed queue is not.
What each surface teaches your hand
There is a physical dimension worth separating from the memory dimension, because the two get conflated in the paper-versus-screen argument.
Writing with a pen involves resistance. The paper pushes back, the ink flows at a rate you cannot rush, and the stroke ends where you lift rather than where you stop moving. That resistance is part of how a hand learns the rhythm of a character, and it is why writing on paper often feels more like the character is being made rather than drawn.
A screen has almost no resistance, and a finger has none at all. Strokes overshoot, corners round off, and the hand learns a slightly different motion from the one a pen would teach. A stylus with reasonable friction narrows the gap considerably.
The practical consequence is narrow rather than sweeping. Screen practice builds the memory and the sequence perfectly well, and it builds a slightly different motor pattern. For most learners that matters only when they pick up a pen after months of screen-only practice and find their characters looser than expected, which a weekly paper session prevents entirely.
None of this argues against screens. It argues for not making either surface the only one, which is the same conclusion the memory side reaches for entirely different reasons.
Common objections, answered honestly
“Writing on a screen is not real writing.” Partly fair for stroke feel and wrong for the memory work. The retrieval, the production of a character from nothing, is identical whether the surface is glass or paper, and that is the part that builds the ability. Handwriting research in Chinese associates writing with literacy development rather than specifying a surface, and a longitudinal study of handwriting fluency is about the act rather than the medium.
“I write better on paper.” Usually true, and usually about stroke quality rather than recall. Check the two separately: can you produce the character at all, and does it look right? A screen is fine for the first question and paper is better for the second.
“A notebook is enough.” It is, if you use it strictly for recall and enforce your own scheduling. Most people do neither, because both are effortful by hand, which is precisely what software removes.
When one surface is clearly right
Four situations where the split does not apply and one surface wins outright.
Preparing for a written exam. Exams are on paper, so the final weeks should be on paper, including timed practice, because writing under time pressure with a real pen is its own skill.
Practising in dead time. Queues, commutes and waiting rooms are screen territory by default, and a session that happens on a phone beats a better session that does not happen at all.
Learning your first characters. Paper first, for a week or two. Getting a feel for how a character sits in its square is faster with a pen, and it makes the screen work more useful afterwards.
Maintaining a large set over years. Screen, because scheduling several hundred characters by hand is the part nobody sustains, and maintenance is exactly where the scheduling advantage compounds.
Practical notes for each
On paper: use squared or grid paper rather than lined, keep a card to cover previous attempts, write the prompt on one side and the attempt on the other, and date each page so comparison is possible.
On a screen: write large, use a stylus if you have one, and turn off any hint or animation that shows the character before your attempt. If your app has a mode where the character fades in as you struggle, treat that as tracing with extra steps.
On both: never look before attempting, always check the order after, and stop the session while you still want to continue.
And keep the two connected. The characters you practise on paper each week should be the ones the screen session has been working on, not a separate list, since the point of the split is two views of the same material rather than two parallel study projects.
{{appName}} handles the screen half deliberately: prompt first, character hidden until you have written it, stroke order checked, misses returned sooner, offline and with no account. It is in early access. The paper half needs no software and is worth keeping, and the daily routine and neat handwriting cover how to structure each.
A week that uses both
Concretely, here is a week that gets what each surface is good for without doubling the time spent.
Monday to Friday, ten minutes on a screen. Yesterday’s failures first, then scheduled reviews, then two or three new characters. Prompt first every time, stroke order checked after each attempt.
Saturday, twenty minutes on paper. Take the characters that passed twice this week and write each one large on a grid, comparing against the printed form for proportion. Then write three or four sentences using them, because sequence and spacing are what single-character practice never rehearses.
Sunday, nothing, or the two-minute screen version if you want to keep the chain unbroken.
Once a fortnight, add a cold test on paper: twenty studied characters written from a prompt with nothing visible. That is the measurement worth trusting, and doing it on paper removes any possibility of a hint.
Total time is under an hour and a half a week, which is a realistic commitment for an adult with a job, and it covers recall, order, scheduling, proportion, stroke feel and writing in sequence.
The record neither surface keeps well
One thing worth adding on top of both, because neither does it by default: a note of why a character failed.
An app records that you got 请 wrong. Paper records the wrong attempt. Neither records that you keep putting the right-hand component too high, which is the piece of information that would actually fix it.
A single line in a notes file per stubborn character, written the moment you diagnose it, is enough: “请, right side sits too high”, “谢, forget the middle stroke”. Reading that list before a session is more valuable than any additional practice minutes, because it converts repeated failure into a specific thing to watch for.
This is also the part most likely to be dropped, since it takes ten seconds and feels like admin. It is the ten seconds that stops the same error from being repeated fifty times.
Frequently asked questions
Is it better to practise Chinese writing on paper or on a screen?
Both, for different halves. A screen hides the character so you practise recall, checks your stroke order and schedules reviews, which are what decide retention. Paper gives better feedback about proportion and stroke quality and leaves a visible record. The arrangement that works for most learners is a short daily screen session and one longer paper session each week.
Does writing on a screen still count as handwriting practice?
For the memory work, yes. Producing a character from nothing is the same retrieval whether the surface is glass or paper, and that retrieval is what builds the ability to write. What a screen reproduces less well is pressure and stroke feel, which affects how your handwriting looks rather than whether you can produce the character at all.
Can I learn to write Chinese with just a notebook?
Yes, if you use it strictly for recall and enforce your own review schedule. Cover previous attempts, write the prompt on one side and your attempt on the other, and keep a list of what to revisit and when. Most people drift back into copying and drop the scheduling, because both are effortful by hand, which is what software removes.
What kind of paper is best for Chinese character practice?
Squared or grid paper, because every character is built inside a square and a grid shows immediately when proportion or centring is off. Lined paper gives you a baseline and nothing about width, which is where most beginner errors live. Write large while learning, since small characters hide the structural mistakes you are practising to find.
What is the best app for the screen half of the practice?
{{appName}} is designed for it: prompt first with the character hidden until you have written it, stroke order checked on your attempt rather than only the finished shape, and missed characters scheduled to return sooner. It runs offline without an account, so the short daily session works anywhere. It is in early access.
