Text-critical wisdom from Gunther Zuntz, Text of the Epistles: A Disquisition upon the Corpus Paulinum (amazon.com):

Readings are either right or wrong. Their theoretical implications largely depend upon this elementary fact. Agreement in genuine readings is normal; possibly significant only where it is confined to few witnesses or where it recurs frequently in well-defined groups; agreement in error suggests some relationship. These theoretical implications apart, we want to know which readings are true. No statistics can tell that: we have got to use our critical faculties and methods. (Zuntz 59)

This is after some paragraphs on the dubious usefulness of “statistical methods” in evaluating and comparing texts/manuscripts where Zuntz has another memorable line:

Variant readings can fruitfully be compared and grouped on more than one principle, but they cannot reasonably be added up or reduced to percentages like the factors of an arithmetical sum. What is the sum total of, say, an egg plus a grape plus a unicorn? (Zuntz 58)

This isn’t a passive, read-before-you-go-to-bed book. Zuntz is a tough slog; you’ve gotta work your mind while reading him in order to get the most from him; consulting the text is necessary too. To grok him more fully I’ll have to read it a few more times (still working on my first slog). But there is a ton of useful stuff in here about the practical application of textual criticism to the real problems one runs into when evaluating variants. I don’t think Zuntz is right about everything, but I do think that if you’ve got any desire to do textual criticism, then Zuntz should be near the top of your list after intros like the Alands Text of the NT and Metzger’s Text of the NT and after you read (really read) the NA27 intro and do some scanning of the consistently-cited witnesses. Zuntz could come before or after Westcott & Hort’s intro (amazon.com) as well. <speaker voice=”yoda”>But read them all you must if textual criticism you desire to practice</speaker>.