Linguistics question: etymologies, and PIE

Browsing Wikipedia, I came across the article on Central Park, NY. Somehow, from looking at Harlem Meer, I ended up at the online etymology dictionary’s entry on the British English word “mere”, denoting a shallow lake, that’s sometimes heard around these parts.

I note that the entry includes a series of similar words in other languages (including the Dutch “meer”), as well as an (hypothesised) etymology stretching back to proto-Indo European (*mari/*mori-).

How are these etymologies developed? Sure, there’s a lot of similar words in European languages, but how, eventually, do they come up with *mari/*mori-? To my untrained eye, it looks like, if we’re being naive about it, that there could be multiple possibilities for a PIE root, so why settle on that particular one?

Further, what is this branch of linguistics called, and is there a book, giving a good overview for a layman, on it?

Here is an excellent online article about reconstruction of Indo-European roots. Scroll down to the section titled “An Example of Reconstruction”, just under a quarter of the way down the page.

It’s called historical linguistics, and the book I’d recommend starting on is The Unfolding of Language by Guy Deutscher. I seem to be recommending this book a lot. This is the third thread I’ve mentioned it in.

One thing to note, as well, is that the asterisk at the beginning of each reconstructed root essentially means “speculatively”. Your untrained eye is not entirely wrong – there are multiple possibilities for these roots. **Mari/mori- are best guesses based on current information and analysis.

As borderlond mentioned, the roots are speculative. The decision is based on probable precursors determined by applying the principles of Historical Linguistics. One of my classmates had a terrible time accepting the fact that this particular science is not an exact one. It really boggled his mind that the reconstructed language might never have existed. The probabilities, though, are that the speculative roots are correct, more likely than not.

(I hate to keep throwing links at the OP, but I don’t find this topic all that easy to distill. My apologies for the forthcoming.)

CRSP, there’s also good information in the Wikipedia article titled “Comparative Method”. The comparative method is the collective name of the steps taken to reconstruct roots of a dead language. The links below lay out the steps in order – seems to me you may be most insterested in the fourth one, “Reconstruct proto-phonemes”:

Assemble potential cognate lists

Establish correspondence sets

Discover which sets are in complementary distribution

Reconstruct proto-phonemes

Examine the reconstructed system typologically

Thanks for the book recommendation and links!