New light from science on how to translate the Bible

From the study (available here):
"The method for answering our question [about the meaning of 2 Cor. 4:1], and for testing our hypothesised answer, involves two kinds of steps. There are steps taken here that are general to scientific method, and steps that are specific to the problem of evaluating a hypothesis about what an utterance in Ancient Greek intended to communicate to sympathetic hearers. The general steps are to disclose the presuppositions, evidence, archive and logic upon which any conclusions are based, together with the investigator’s perspective (i.e. likely sources of bias), and to attempt to falsify any hypothesized answers to the research question. This is a reflexive, inductive-deductive method. One or more researchers work in a self-aware way (reflexively), to found a hypothesis on evidence (induction), then to draw out falsifiable implications of this hypothesis (deduction), which are then tested."

"There are at least two tests that a hypothesis about the intended meaning of an Ancient Greek utterance ought usually to face, although neither is decisive:

(1) Does it increase, relative to alternative hypotheses, the coherence of the text to which it belongs? Coherence of text can be defined as the extent of informational relations between the words (and for example punctuation) of the text.[1] For example, consider what informational relations, and therefore coherence, one can identify in the following texts: (a) He loves God; (b) We loves God; (c) They we he. Interesting, possibly useful descriptions of different kinds of informational relations have been developed over at least three decades by scientists working with Linguistics and Artificial Intelligence,[2] and in this essay we will especially consider two kinds: cause-effect and elaboration.

(2) Does it violate reliable real-world knowledge?

The reasons that neither is decisive are, firstly, that some texts are incoherent, and, secondly, that real-world knowledge is never perfectly reliable.[3] Nonetheless, these tests are usually inescapable, and optimal starting points for evidence gathering"

[1] This is a proposal I am making. Among related alternative proposals (cf. de Beaugrande & Dressler 1981, 48-112), its closest relative is the proposal by Wolf & Gibson 2006, 1-2.

[2] Cf. Wolf & Gibson 2006; Renkema 2009.

[3] Socrates/Plato, the apostle Paul (1 Cor 13), David Hume and Karl Popper all argued for this limitation.