Update: I've left my original two opening paragraphs here (slightly modified for contextual clarity) because I still believe in general they are true when it comes to resolving highly disputed variants in the text. Often the reason they are highly disputed is because the extant textual witness cannot answer which variation is correct in a straightforward manner (i.e. in a manner acceptable to both methodologies of approaching textual criticism). However, further discussion after the question revision to clarify what the OP was getting at (see below) has opened up the theoretical possibility of proving the necessary inclusion of the phrase based on purely textual grounds. More evidence would be needed to determine if (1) that possibility is in reality not open (by finding other valid instances of the construction), which leaves the variants unresolved on textual grounds, or (2) it does in fact resolve to answer in favor of including the phrase.
Original Answer
In this case, you cannot "conclude" anything from "the text in Greek" itself. Some manuscripts have it, some do not.
Rarely will "the characteristics of the Koine Greek" answer whether words should be included or not (such characteristics are mainly valuable in examining word spelling or order variations, which in Greek often [though not always] do not matter anyway).
The only way one can resolve this for themselves is at the presuppositional level. Consider this summary analysis of the passage (bracketed comment is mine, and emphasis added):
The words ἐν Ἐφέσῳ are absent from several important witnesses (P46 א*
B* 424c 1739) [five textual manuscripts] as well as from manuscripts mentioned by Basil and the
text used by Origen. Certain internal features of the letter as well
as Marcion’s designation of the epistle as “To the Laodiceans” and the
absence in Tertullian and Ephraem of an explicit quotation of the
words ἐν Ἐφέσῳ have led many commentators to suggest that the letter
was intended as an encyclical, copies being sent to various churches,
of which that at Ephesus was chief. Since the letter has been
traditionally known as “To the Ephesians,” and since all witnesses
except those mentioned above include the words ἐν Ἐφέσῳ, the Committee
decided to retain them, but enclosed within square brackets.1
This is a good example of one's presuppositional position having to answer the question, namely
- Do you believe the small number of "older manuscripts" necessarily (or at least are more likely) to contain the correct reading (OMITTED). Especially in light of the fact that as best I can tell on a quick look, there are still old witnesses (P92, A, C) that do have it.
- Do you believe the larger number of "majority manuscripts" (of which there are many, hundreds at least... I do not know the exact count for Eph 1:1) necessarily (or at least are more likely) to contain the correct reading (INCLUDED).
Essentially, if you believe God would be at work to preserve His word through insuring a large, accepted and recopied distribution (i.e. majority text witness), then you are going to believe it should be included. If you are going to believe God was simply at work to preserve at least a few witnesses to His word, then you are at least open to it being omitted based off some exclusion (i.e. partial ancient witness). One's presupposition on preservation drives the answer.
The issues in textual criticism are the fact that ancient witnesses only tell us what they record, not what other witnesses may have recorded which we no longer have to examine. So they can only prove their reading did exist at that time, but cannot disprove that another reading (to which we have later witness) did not also exist at that time.
As the quote above noted, and as you saw in your quote from NA28, the text is kept in brackets because they did not feel there was enough evidence to exclude it, but in their way of looking at the text, there was enough evidence to question it. Usually if you see text in brackets, you can bet the majority witness testifies to it, but some of the older witnesses do not.
For the record, I hold to essentially a majority witness position.
Answer to Question in Comment/Updated Question
Assuming the phrase was not in the original, then the translation of the Greek would be (literal word for word order here):
Παῦλος ἀπόστολος Χριστοῦ Ἰησοῦ διὰ θελήματος θεοῦ τοῖς ἁγίοις τοῖς
Paul, apostle of Christ Jesus by [the] will of God, to the saints, the
οὖσιν καὶ πιστοῖς ἐν Χριστῷ Ἰησοῦ
ones who are also [or even] faithful in Christ Jesus.
The inclusion of καὶ (kai) in such a case, the conjunction normally translated "and," but that can have other meanings such as "also" or "even," does not definitively (see next full paragraph) eliminate the possibility of the omission, but does possibly offer additional support for the majority witness reading that would take it as "and." The above translation seems "odd" to me only because either:
- What other "saints" would there be than those faithful to Christ? It does not add anything in defining a "saint," especially in a salutation.
- Would Paul only be specifically addressing "saints" who were especially faithful to Christ? The context of the epistle would seem to not exclude any "believer" by addressing only "super believers" (so to speak), so this also seems unlikely.
I have to admit not having done a study to find out if this is a construction attested to anywhere else (in Koine or otherwise... if such a study has not been done, that could be fodder for a PhD dissertation, as it would be a laborious study). So I may have to reverse my opening comments about whether the Greek text itself or "characteristics of the Koine Greek" may help resolve it. If a study were done that could demonstrate καὶ never follows a participle of being verb (like οὖσιν) except in a conjunctive aspect "and" relation (as opposed to an additive "also" or intensifying "even" function), then that would be proof the omission is invalid. Even if it were found to very rarely be used such, the use cases themselves (the conditions of the context of the other places) may also validate or not the inclusion.
Notes:
1 Comment is on Eph 1:1 by Bruce Manning Metzger, United Bible Societies. A Textual Commentary on the Greek New Testament, Second Edition a Companion Volume to the United Bible Societies’ Greek New Testament (4th Rev. Ed.). London; New York: United Bible Societies, 1994.
