All throughout human history, the question about the origin and evolution of (human) language has exerted a fascinating appeal to both scholars and laymen. Thanks to the progress in computational methods, it has been possible in recent years to attain new insights into this subject matter and to conduct simulations with the pretence of verifying or falsifying some of the assertions proclaimed by an earlier linguistic and philosophic tradition. By means of a simulation model made up of populations of artificial speakers, this work sets out to address the problem of language change and its variation over time. Under what conditions can -- and do -- languages change? More fundamentally, it is asked under which circumstances languages are learnable, and whether the assumption of a particular human qualification for language ability is really necessary.