[int_19h]

In this case, it is, nevertheless, correct. There's simply no clear-cut objective boundary here. You can arbitrarily define one as N% of mutual intelligibility, but then your pick of N is itself a political question.

Keep in mind that many of what we today consider distinct languages historically formed a dialect continuum - usually geographic, where you can observe gradual changes as you travel, which may add up to the extent that there's no real mutual intelligibility between the beginning and the end of your path. Are they different languages at that point? If so, what does that make of all the gradual steps in between?

The stark language boundaries that we're more familiar with today are largely due to states aggressively encouraging (or outright forcing) some kind of language standard on what was before a dialect continuum area, smoothing out the differences within it while sharpening them at the borders. This process kicked into high gear with the rise of nation-states, which would often do so for ideological reasons - basically, to force conformity and reify the abstract notion of "one people". But it hasn't happened everywhere to the same extent as it did in Europe, and there are still plenty of places around the globe where clear language boundaries don't really exist.