Evidence that most Indo-European reconstructions are artefacts of the linguistic method of analysis

.A typical book on Indo-European linguistics takes for granted that the evidence for the Indo-European family tree is ‘obvious’ or ‘compelling’. In more specialist research books, one finds a focus on how to ‘fit’ the data to the model. The earliest example of this was Verner, who famously described ‘exceptions’ to Grimm’s Law. Verner modified the model by introducing ‘contextual specifications’, thus converting the exceptions into ‘apparent exceptions’. Over the years, the counter-evidence to the model has been ‘explained away’ with similar processes, so that today one can find very little direct counter-evidence to the model. Here I examine this process critically. I ask if the linguistic method has become circular – in other words, I ask “Have we fitted the data to the model, or have we made the model so flexible that it can fit almost any data?