I can think of two strong arguments against picking a less-discussed (dying?) language:
1) Anecdotally, the top 1% of programmers (the people you really want to work with) will go to the tools that make them the most productive. The reason there's so much hype around languages like Ruby is that it makes programmers feel more productive and happier. The kind of programmers that go online and talk about these things after hours and argue about minor semantics are the passionate ones, and the most likely ones to rock from 9-5.
A 2% increase in productivity matters to a programmer that already gets twice as much done as an average code-monkey.
2) One also has to look at what languages are being learned by new coders. Code today might need to be maintained in 5, 10, or 20 years (let's hope not...). If Perl coders are sparse now, what happens when you need to hire a contractor at 2am in 10 years? You don't want to pay for one with an obscure skill-set.
I'm a firm believer in choosing the best tool for the job, and not jumping on the 'newest thing' every time a new blub gets an endorsement from this year's DHH, but; ceteris paribus, I'd pick a language that's gaining momentum in the world, not losing it.