The end-to-end arguments have been at the heart of an increasingly political debate about the proper place of communications functions in a physically and administratively distributed network such as the Internet. In particular, what is the right balance between application end points on the one hand, and network intermediaries on the other? This thesis goes beyond the cursory arguments found in much of today's discussions on Internet governance and regulation. It unveils the historical background to the end-to-end principle and develops them with a rigour that has thus far been absent from the discourse on Internet design principles. Our thorough and timely elaboration of one of the core technical principles that came to form today's Internet is relevant to most anyone with a stake in the future Internet evolution. If we are to maintain the goodness of the Internet, it is vital that we first understand its principles. Only then can we hope to arrive at useful policy implications.