The writing system is the smallest part of a hard language

Ask someone why they have not started Russian and they will say the alphabet. Ask about Korean and they will say the characters — usually not knowing that Korean does not use characters. Ask about Japanese and they will say all three systems at once.

The writing system is the visible obstacle. It is almost never the expensive one.

The arithmetic, and why it is robust

Here is the awkward thing about making this argument carefully: there is no authoritative published figure for how long a script takes on its own. FSI publishes whole-language hours and nothing script-by-script, so any specific number would be somebody's guess — including mine. We have a writing-system tool listed as blocked for exactly this reason.

But the conclusion does not need a precise figure, because it survives any plausible one.

Take Korean, at 2,200 FSI class hours. Suppose learning hangul takes a generous forty hours — several times longer than most learners report, chosen deliberately to make the argument harder.

Forty hours is 1.8% of the total.

Now be more generous still and say a hundred hours. That is 4.5%. Take Russian at 1,100 hours and allow eighty hours for Cyrillic: 7%.

The precise number does not matter, and that is the point. Any figure you can defend leaves the writing system a small single-digit percentage of the language. The remaining nine tenths is grammar, vocabulary, listening, and producing correct sentences at speed.

Chinese and Japanese are the real exception

Two languages break the pattern, and it is worth being exact about why.

Chinese characters are not an alphabet. They are a vocabulary-sized learning task that never stops — thousands of items, each with a form, a sound and a meaning, acquired continuously alongside the language rather than in a burst at the start. That is a substantial fraction of Chinese, not a rounding error.

Japanese runs three systems at once, and its kanji carry multiple readings that depend on context, which Chinese characters overwhelmingly do not. Japanese is also the only language in FSI's hardest band carrying its "usually more difficult" marking.

But notice what separates these two from Korean, Russian, Greek and Arabic: it is not that their scripts are harder to learn. It is that their scripts are not a fixed-size task at all. An alphabet is a closed set you finish. Characters are an open set you accumulate.

That distinction matters far more than "hard script versus easy script".

Why people get this backwards

Because the writing system is the only part of a language you can see before you start.

You cannot look at Russian aspect, or Korean honorifics, or the Arabic verb system and feel how much work they are. You look at Cyrillic and feel foreign immediately. The feeling is real and it is not proportional to the cost.

The practical damage runs in both directions. People avoid languages over an alphabet that would cost them a fortnight. And people who do start treat finishing the alphabet as major progress, when it was the cheap part — which sets up the discouragement that arrives a few months later, when the actual work begins and does not have a visible finish line.

What to do instead

Learn the script first and quickly, then stop thinking about it. For an alphabetic system, do it in the first fortnight and move on. It is the only part of the language with a definite end.

Do not choose a language on its alphabet. The hours table is the number that should drive that decision, and Greek and Russian sit in different bands despite both having an unfamiliar alphabet.

For Chinese and Japanese, budget characters as vocabulary, because that is what they are — continuous, never finished, and part of the 2,200 rather than a prelude to it.

What the hours actually go on, language by language.