Between late 1965 and mid-1966, following a failed coup blamed on the Indonesian Communist Party, an army-led campaign of mass killing swept the country. Estimates of the dead run from several hundred thousand to around a million, with hundreds of thousands more imprisoned for years without trial. Ethnic Chinese Indonesians, associated in propaganda with communism and with the People's Republic, were among those targeted, unevenly by region but severely in some.
What followed
Out of that period came Suharto's New Order, which governed Indonesia until 1998, and with it a formal policy of assimilation aimed at the Chinese population. Public expression of Chinese religion, belief and custom was restricted by presidential instruction. Chinese-language schools were closed and Chinese-language publishing all but ended. Ethnic Chinese citizens were pressed to adopt Indonesian-sounding names, and identity documents recorded their descent in ways that followed people through every official transaction of their lives. Celebrations that had been public became private. Temples continued, more quietly.
The restrictions were dismantled after the New Order fell: the presidential instruction was revoked in 2000, and Chinese New Year became a public holiday shortly after. That is recent. For the whole of the period during which Chang is said to have trained, taught, and been filmed, the environment for a Chinese cultural lineage in Java was one in which visibility carried real risk.
What this explains, and what it does not
It explains a great deal about the shape of the evidence. It explains why a Chinese teacher in Java would restrict transmission to the Chinese community, why no records were kept, why a practitioner would decline to be identified on television, and why the caution in every account of this tradition reads as something other than showmanship. When Chang asked the Blair brothers not to name him and later asked for the footage to be withdrawn, he was doing something with an obvious cost and an obvious reason, in a country where the reason was current.
It does not, of course, establish anything about the practice. Political conditions can explain missing documentation without supplying the documentation, and an archive that treated persecution as evidence would be arguing rather than recording. The correct statement is narrow: the absence of a record here is what the period predicts, so it counts less against the claim than the same absence would in an open setting — and no more than that.