The thing is that people curing cancer (or doing any other kind of research) famously don't spend their time learning niche languages. Quite frequently they can't program at all. If they can, they usually write some really horrible code in whatever language they learned at school -- mainly because they are too busy doing their research to care about programming.
Programmers, on the other hand, do spend their time learning languages (at least some of them, anyway). People frequently ask me how many programming languages I know. I've lost count of the programming languages I've used on commercial projects (it's well over 20). I don't "know" any of them any more. I stupidly put C++ on my CV the last time I was looking for a job and someone started asking me about Boost. I had to reply, "The STL had just been released the last time I used C++. We thought RogueWave was pretty cool at the time". I mean it's actually a completely different language right now. I need to have a section on my CV for "Can tell you amusing anecdotes for this technology but can't actually use it any more".
However, the important thing is that if you throw tens of thousands of lines of legacy code in any language that is in a style I'm familiar with, I'll be up to speed in a week or two. That's because I'm a professional programmer. It's what I specialise in.
New programming languages are great for a number of reasons. First it moves the state of the art forward without having to be backward compatible. Yeah, the state of the art moves slowly and each new language only adds a tiny bit of improvement, but over time it adds up. You don't want to be using CFront like I was at the beginning of my career. Pretty much any modern language you pick is going to be head and shoulders above that.
But the other thing is that becoming fluent in new languages makes you a better programmer in every language. You see idioms you've never seen before. Techniques that are the bread and butter in one language are unheard of in another language -- mainly because you've got crusty old guys like me who've done the same damn thing every day for 30 or 40 years.
I've actually worked in a physics lab before. They wrote some important papers during the time I worked there and I helped -- by writing the code they needed to analyse their data. That was when I was very young (20???). I'd do a much better job of helping them now, in part because I'd do a better job of taking their hacked up Fortran code and making something that could be maintained long term.
Yeah, you don't need to be a fanboi for some language or other, but you also don't need to be a luddite either. Times change and new things happen. It's one of the fun parts of being a programmer.