I plotted the distribution of the numbers and got a nice half-bell shaped curve, making me think it was the first occurence of a character in some text. Since I am Swedish, I know very little of Lincoln and started with the Gettysburg Address.
It turns out the transcriptions of the speech varies a bit. The versions seem to differ whether they spell "Fourscore" or "Four score" as well as commas here and there. In any case, the first four letters seemed to be invariant. Hence:
1-F, 2-o, 4-r => _ _ o _ _ _ o F _ r
Next, I wrote a Python script going through the text printing the first occurence of every letter. I tried both including spaces, periods, and commas, as well as excluding them. However, the first occurences when excluding these characters did not line up nicely with the given numbers.
In the transcript with all characters, only one letter was close to place 103, namely
L, which had its first occurence at place 107 in my transcription, we now have _ L o _ _ _ o F _ r
If picking a version spelling "Four score" using commas in the first paragraph, such as this, all other numbers line up exactly with the first occurrence of a letter. The resulting text is
iLovegoFer -- I love gofer
Could this be correct?