I was wondering, what is the method (or the methods) that linguists adopt to understand and know the etymology of a word?
Are these methods reliable and in what measure?
The knowledge I have on the matter is not that deep so I wanted to know exactly what happens "behind" that Origin entry in my dictionary.
As you can probably guess, there are several ways to determine the etymology of words. Probably the most reliable is to look at the genetic relationships between languages - i.e. we know that a word comes from Latin or Greek because it is preserved in a similar form across several modern languages, and we have historical documents in the original language that also use this word (or a very similar form).
Often, a historical phonological process such as I-mutation (umlaut) is observed for a group of words, which can help to prove that they came from the same language.
The etymology of a word becomes more speculative the further back we go - we can be pretty sure that a word came from Sanskrit, Arabic, Latin, or Greek, because these languages are (relatively) modern and well-documented. However, determining which proto-language a word is descended from can be more controversial.
In summary, historical documentation and phonological similarity are the two main ways to determine a word's etymology.
Most English words have roots in Indo-European, either through original Germanic or through borrowed Latin, French, or Greek words. For all of the 19th Century, every linguist in the world (which principally meant Europe, in this case) was busily engaged in bringing Ordnung to the prehistory of the Indo-European languages.
In the century since, this immense scholarly undertaking has been codified and turned into individual etymologies, which are normally looked up in a principal source, like Pokorny's Indo-European Etymological Dictionary. To make sense of these, you need some linguistic training.
There are secondary sources available that skip a lot of details but still do the job.
Buck's monumental Dictionary of Selected Synonyms in the Principal Indo-European Languages: A Contribution to the History of Ideas is now available in paperback. It's organized like a thesaurus, in 21 chapters by topic (4. Parts of the Body; 8. Agriculture, Vegetation; 15. Sense Perception), and in each chapter about a hundred or two ideas -- like head (4.20), rose (8.58), restore (11.23), or good smelling, fragrant (15.25) -- are given with a list of synonyms and a discussion of the roots involved and the history of the words and concept in different languages. This is a fascinating book to dip into anywhere; Buck really meant his subtitle -- the first sentence in his Preface is "How do we get our ideas?"
Watkins' American Heritage Dictionary of Indo-European Roots was compiled specifically for the words in the unabridged American Heritage dictionary, but they cite Pokorny and have illustrative examples.
