Evolution is the change in the heritable characteristics of populations over successive generations. It is the dominant paradigm in biology and it explains how species get from A to B and then to C through mutation and genetic recombination. And through other stuff, too.
Before the invention of the printing press, texts evolved similarly. Text T1 was composed by author A, who was also scribe S1, writing it by hand. In the next generation, scribe S2 comes along, endeavouring to copy T1, but making inadvertent or deliberate changes to it, through either unavoidable error (fatigue, eyestrain, carelessness) or by tampering with it. T1 changes under the scribe’s quill to become T2.
T2 meets scribe S3 in the next generation, who makes further changes, and so on. The process is unavoidable because errors in human-based reproduction tend towards 100%, while errors in mechanical replication like print or digital lean towards 0%. Each generation has a scribe, each scribe makes changes, even when they want to act like a machine, which brings in more and more mutations to the original T1. By T100, the text may be as different from T1 as a mammoth is from a fish.
Occasionally, a scribe comes along wearing a T-shirt (which is short for Textual-shirt, of course) that says scholar on it. They set out to reverse the process, by collecting multiple Ts from previous generations and making new changes to new Ts. Like genetic engineering, this has its benefits and its pitfalls. At best, it minimizes error; at worst, it creates even more noise, obstructing the visibility of other Ts. There are good scholars, and there are bad ones.
Philology is to texts what genetics is to biology. One of its main tasks is to recover the DNA of a text by tracing its evolution through the generations back to its original incarnation, from T100 to T1, and then to author A, who by T100 may be a faded memory.
The problem is this: subsequent Ts are to T1 what individuals are to species, and like biological individuals, Ts may disappear without a single trace. By this point, you will have guessed that Ts are texts in manuscript, and their survival tends towards 0% through history, especially when they encounter political and social instability, cultural and technological change, fashion shift, etc.
Philology, armed with several other sub-disciplines about which philologists are constantly squabbling, has another duty: that is to bring back those Ts which are no longer around, by looking at earlier and later Ts and figuring out what lost Ts would have been like. The objective is reverse engineering – going back to T1 even when T30 and T50 are missing. The result is a hypothesis, of course, just like the recreation of a lost animal species of which only later mutated individuals survive.
There are no definitive answers, just a constant struggle towards 0% error.