The word perfect is not extant in the original manuscript, but spacing between extant fragments indicates that a couple more letters could have been there. That possibility suggests two emendations here in Alma 42:15: (1) “a perfectly just God”, and (2) “a perfect & just God”. One could argue that as Oliver Cowdery was copying the text from 𝓞 into 𝓟, he either accidentally dropped the ly at the end of perfect or he skipped the ampersand. Of course, the reading in 𝓟 of perfect just can be interpreted as a case of asyndetic coordination of two adjectives. The 1830 compositor placed a comma between the two words, thus treating each one as an adjective; all the subsequent editions have followed this conjunctive interpretation of perfect just.
The reading of the current text does seem rather strange. First of all, there is nothing to suggest in the larger passage that God’s perfection is at issue. Secondly, if the original text actually reads “a perfectly just God”, then there is a balanced chiasmus here in Alma 42:15:
a to bring about the plan of mercy
b to appease the demands of justice
bª that God might be a perfectly just God aª and a merciful God also
In other words, the atonement allows God to be both “perfectly just” and “merciful”.
From a stylistic point of view, there are no other examples in the text of perfect being used as in “a perfect just God”. We do have examples of the adverbial perfectly modifying a following adjective:
These two examples show quite clearly that “a perfectly just God” is a possible emendation. On the other hand, there is an example of perfect conjoined with another adjective but, we should note, with a separating and:
This latter example supports the emendation “a perfect and just God”.
There is some minor evidence in the manuscripts for the scribes having difficulty with the adverbial -ly ending, as in these two examples in 𝓟 where Oliver Cowdery was the scribe:
And there is considerable evidence that the scribes sometimes omitted and (typically written as an ampersand). For some examples of Oliver’s occasional omission of and, see the joint discussion under 1 Nephi 17:39–40 and 1 Nephi 17:40. So there is manuscript evidence for the possible loss of the -ly ending or of an ampersand here in Alma 42:15.
One final possibility worth considering is that perfect is actually an adverb form but without the -ly ending. Such adverbial forms can be referred to as bare adverbs; Merriam-Webster’s Dictionary of English Usage refers to them as flat adverbs (see the discussion under that name). In general, the original Book of Mormon text allows for the bare adverb, as in the following examples from the earliest sources:
But all these adverbs modify a verb rather than an adjective. The only obvious case in the original text of the bare adverb modifying an adjective is the ubiquitous exceeding, as in 1 Nephi 2:16: “I Nephi / being exceeding young”. The 1920 and 1981 LDS editions have grammatically emended the many examples of adverbial exceeding to exceedingly (for discussion, see under 1 Nephi 2:16 as well as under exceeding in volume 3). There is only one other case where an adjective has been interpreted as a bare adverb (namely, in the LDS text since 1920):
For discussion of the evidence that wonderful in Alma 51:11 may actually be an adjective, not an adverb, see under that passage. Given usage elsewhere in the Book of Mormon text, there is little evidence, apart from exceeding, that perfect in Alma 42:15 should be interpreted as a bare adverb. There are a few instances of perfect just on Literature Online which suggest the possibility that perfect just itself is correct and that it means ‘perfectly just’. The following two examples of perfect just come from the late 1800s and seem to mean ‘perfectly just’ (here accidentals are left unchanged):
Of course, these examples considerably postdate the Book of Mormon. But they should make us hesitate before emending perfect just. We should also note that the lacuna in 𝓞 is sufficiently large that we cannot insist that the additional space must have been filled by the very short ly or &. Nor can we confidently exclude even the asyndetic reading for perfect just (that is, with the meaning ‘perfect and just’). In fact, one important fact to consider is that for this part of the text the 1830 signature was proofed against 𝓞, from Alma 41:8 through Alma 46:30 (see the discussion under Alma 42:31), and in this instance the 1830 edition ended up reading perfect just, which suggests that 𝓞 itself read this way (but, of course, a minor difference like this could have been missed in proofing). In any event, the safest solution here in Alma 42:15 will be to retain perfect just, the reading of all the printed editions as well as the earliest extant source, the printer’s manuscript.
Summary: Maintain in Alma 42:15 the difficult reading “a perfect just God”; although this reading is difficult, it could represent the original reading, with either the bare adverbial meaning ‘a perfectly just God’ or the conjunctive meaning ‘a perfect and just God’.