Buttmann’s intermediate Greek grammar was translated into English twice, once in Boston and once in London. And while the first translation was an abridgment, the contents of the two translations in the section on tense differ little. For that reason, the larger edition is the focus here which, ideally, more accurately represents the contents of the German original. I have compared the German original and English in small sections and they appear to be in strong agreement.
The section on tense begins with the rather odd statement,
With such a dismissal of the other forms, one can only hope that students in the 19th century were not using Winer and Buttmann side by side, since Winer’s grammar was intended to only cover the differences of the New Testament from the rest of the language and Buttmann covers very little of just that. There are only a couple other English grammars that one could use, that of August Matthiä and Friedrich Thiersch, both of which are available in English translation will be discussed in due time.
As for his actual discussion of the aorist, Buttmann makes a number of comparisons with other past tenses. Methodologically, then, he views it as a necessity that a given tense can only be understood in light of the larger system. This also gives us some insight into how he views other forms. The imperfect, perfect and pluperfect are all preterites for him. The imperfect continues in a similar vein as Winer, being used “to state the circumstances, by which the thing, which happened, was attended, when it happened.” This is generally comparable Winer’s secondary/subsidiary conception of the imperfect.
The basic distinction, for Buttmann between the aorist and the perfect also arises from discourse actors: unlike the aorist, the perfect does not narrate. Instead the perfect “connects what has happened, as past, with the present time.” This description is closer to that of Wallace and the tradition he follows than it is to idea of the perfect as stative. But there are a couple key clarifications that follow. Buttmann comments in an accompanying footnote:
“It will always be found that the pure perf., such as it has particularly maintained itself in Greek, is used only when the consequence of the performed action, or even of its ceasing are still connected with the present time. He who says, I have known it, says at the same time, I do not know it any longer. He who says, οἶκον ᾠκοδόμηκα, conveys the idea of the house is being still standing; but if he says, ᾠκοδόμησα, he leaves it at least undecided, and he uses the same expression, when he positively knows that the house is no longer standing.
There are a couple of ambiguities here. First, it is not clear exactly what Buttmann means by the pure perfect and whether it should be distinguished from the perfect of the main text. One explanation might be that pure refers to the use of the perfect historically. If that is the case, then minimally, Buttmann’s words are a grammatical version of the etymological fallacy.
But even if that’s the case, it does not supplant the significance of the description particularly in light of his qualification: in as much as “it has maintained itself in Greek.” To the extent that the language has not changed dramatically, the statement continues to be true. We might then interpret Buttmann’s words as recognizing the existence of language variation and change within the perfect, a change that eventually saw the disappearance of the perfect from the language as its semantics became difficult to distinguish from the aorist. If this reading of Buttmann is accurate, one must wonder if the current fad of seeking unified explanations of grammatical features at the expense of variation and change is somewhat wrong-headed.
Buttmann goes on to note that the aorist can be used for the perfect, an observation which we might parallel with the contemporary linguistic concept of asymmetrical markedness. Specifically, Buttmann has already stated that the aorist (unlike the perfect) leaves undecided (=unmarked) whether a given state or condition continues to exist in the present, while the perfect necessitates that it is (=marked). That the aorist can be used where a perfect can be, is merely evidence of this fact in practice. Buttmann’s observation here might be useful in examining those examples where scholars have claimed that the perfect is merging with the aorist in the Hellenistic Koine or even as early as the Classical period. Do those supposed examples involve perfects expressing a state of affairs that continues to hold to the time of speaking? If so, then, it is unlikely that those are aoristic perfects.
 The specific time of this final change is debated. Evans argues for the Byzantine period (Verbal Syntax, 145-146), while Caragounis holds to a much earlier date (Development of Greek, 154), though he does little more than regurgitate Jannaris (Historical Greek Grammar, 439) without any actual discussion of his own on the matter.
 This is perhaps most represented by Porter (Verbal Aspect) and Campbell (Verbal Aspect), whose claims for the Hellenistic period find little foundation in the historical development of the language.