Theories on the origin of Bezae’s text are not in short supply. After surveying no less than nine theories on the origins and development of the “Western” text, B. M. Metzger remarked, “one is impressed by the wide diversity of hypotheses and the lack of any generally accepted explanation.” 
Consolidating these nine theories into four related groups, we have:
- Bezae’s text (of Acts) is attributable to different editions (of Acts), theories 1-4
- Bezae’s text reflects the editorial and/or evolutionary development of a free text after the mid-second century, theories 5-6
- Bezae’s text is representative of the earliest-attainable text form from before the mid-second century, theory 7
- Bezae’s text is assimilated to one of the old versions, either Latin or Syriac, theories 8-9
What is striking (and perhaps revealing) about these groups of theories is their mutual incompatibility. Obviously, Bezae’s text form is either earlier, later, or roughly contemporaneous to the mainstream alternative. But it can only be one of these! Likewise Bezae’s distinctive tradition is either initially Greek or initially Latin/Syriac. But again, it is unlikely to be both.
Yet faced with these main theories and their numerous variations, how are we to decide between them? How do we test a theory of Bezae’s text?
It seems worthwhile at the outset to acknowledge the historical nature of the inquiry. Any theory of Bezae’s origins makes historical claims that bind smaller observations to a larger framework, forcing us to draw connections outside the text itself and outside the textual tradition to the period literature. A textual theory consists of more than noting agreements between witnesses.
A good test discourages theories that are merely plausible, while drawing few connections beyond the text itself. At the same time, a good test encourages comprehensiveness in accounting for the full range of observed features in both text and manuscript. A theory is more likely to be accurate — and useful — if it is comprehensive in its integration of the data under the proposed framework. While critics will inevitably differ on matters of interpretation, it seems they should agree that the key evidence is addressed.
It is inevitable that some periods are better documented than others. Yet we should expect the evidence to be well-integrated with its proposed historical background to the level of detail permitted by the sources. At the same time, the theory’s claims should be corroborated by a range of independent sources and diverse lines of evidence. A robust theory will absorb new evidence and accommodate or even suggest new lines of inquiry.
But it is evident that many of the proposed theories of Bezae’s text do not meet these criteria. For example, multiple-edition theories are incomplete from the outset because they only address the evidence of Acts, while overlooking the four gospels. But if it is thought that the gospels do not belong to the same tradition as Acts, this must be defended. An account must be given of the gospels not merely of Acts.
We might also consider theories of Bezan priority. While such theories claim a historical framework within the limitations of second-century evidence, they have difficulty explaining the local genealogical evidence suggesting that many Bezan readings are derivative. This weakness in turn raises suspicions regarding the limitations of the historical data. It now appears that the theory survives on gaps in our knowledge.
Clearly then, we are not lacking in the supply of theories of Bezae’s text. Yet we must wonder whether the multiplicity of theories disguises a fact that none yet tells the whole story.
 Bruce M. Metzger, A Textual Commentary on the Greek New Testament (2d ed.; Stuttgart: Deutsche Bibelgesellschaft, 1994) 233.