Sure it can. Just run all possibilities against a dictionary of all words. It’s Trivial.
No, Okie. You don't know what you are talking about. A dictionary will not find a pad's coding for page, line, word, character location in word. It's impossible. That data will change for EVERY SINGLE LETTER IN THE CODE. It changes for every letter, space, number, period, comma, colon, apostrophe, etc. There is no way on earth that anyone can tell what code sequence is which character. No single word is ever represented. All the decoder has is a series of numbers and may not even know where a specific character definition starts and stops. There may be nonsense numbers between characters for all he knows. I would set it up that way. That's what is trivial if you program the devices to handle it all.
Here's an example using just one reply from one thread and one reply from an unknown FreeRepublic Thread as the one-time pad:
BEGIN
16302902120303003020303103090103203010203302190303404050
2035040104036020505037031702038020710039032103040046100504102130
304202160604303100104404100204504070504110204603220304703110104
802130704904110305002090205102130905201040205301010105403130
4055010509056011102057021905
END
(line breaks are purely random and represent nothing pertinent to the code or the sense of the message. Consider the code as one continuous line.)
I will tell you that your name, Okie, occurs once, and the string "code" may occur more than once in this encryption. I am using a one-time pad. Someone who knows the key and the one-time pad thread could easily decipher it. I took some time and honestly constructed this message. However, no amount of brute force with a dictionary will ever make a dent in trying to decipher it. . . because not one letter, character, punctuation mark, word, etc. is repeated in the encryption above, nor do any of the code groups (if there are any) represent any particular words that might be found any any language's dictionary.