: Sure, it's only one word (that I found).
: However, you can't just say that it's "just
: a coincidence". Of any other language, was
: Japanese and not any other language?
Since you're claiming there is a relationship
betweeen Cantonese hai and Japanese hai, you're
the one that has to show that it is the case.
I presume you are claiming that Cantonese hai
comes from Japanese hai, right?
There are several problems. One is that
Japanese h- used to be p-. When they borrowed
words such as 'sign' from Middle Chinese
around the time of the Sui and Tang dynasties,
(paai in Cantonese), they first rendered it
as *pai, but then p- became h- in Japanese
(you can see this in places like the kana system,
where the non-"voiced" version of ba/bi/bu/be/bo
is no longer *pa/pi/pu/pe/po like it was when it
was set up, but ha/hi/hu/he/ho). However, p-
came into the language later, so when they
borrowed 'sign' again (in the sense of 'mahjong
tile', it was borrowed as pai, and is so to
this day. If you're proposing that Cantonese
hai is from Japanese, then you need to show
that such a word existed in Japanese at that
point in history (pre-Qin times?), and also
explain how and why *pai become hai, since
Chinese languages have always had p-.
Also, hai 'yes' in Cantonese is really 'it is',
i.e., the copula, rather than a special word
just for "yes". (Compare its antonym,
m-hai 'no', which is actually 'it is not'). This
is not the case in Japanese, where the copula 'is'
is not hai, but da/desu (and various other similar
variations). However, they do have hai/ee as a
separate word for 'yes' (and iie/ie for 'no').