Any English-speaking student learning "Classical" Nahuatl needs this book; there is currently nothing remotely comparable in English treating the language of the two centuries following the Spanish invasion. Hippocrene has published a "Nahuatl-English/English-Nahuatl Dictionary" but this covers only the modern language. Other works, such as Campbell's redaction of Molina, not to mention Simeon (in French) are hard to come by outside of major universities.
However, the student will quickly discover some serious drawbacks in this work. First off, by "analytical" Ms. Karttunen means that she broken down the sentences and phrases glossed in Molina's great Spanish-Nahuatl/Nahuatl-Spanish lexicon into what we would call "words". Some bound morphemes are presented under their own heading as well, but not all. Nahuatl is characterized by a fairly small phoneme inventory and quite a lot of assimilation takes place, so further analysis would have been useful, at least for those of us studying on our own.
A more serious problem is the restriction on sources. Ms. Karttunen started from Molina, as anyone would, and has added material from other vocabularies, including a couple of modern ones. All to the good, but it appears to me that she did not much go beyond this. The "Bancroft Dialogues" were utilized (Ms. Karttunen cooperated with James Lockhart in producing an excellent edition of this text) but this is high-flown language and not an obvious candidate for inclusion in a basic lexicon. I have been working recently with Sahagun's great work, the "Historia General De Las Cosas De Nueva Espanya", a landmark in anthropology, folklore, and sociology (though no literary masterpiece) which any student of Nahuatl will want to sample, and so far I see no evidence that Ms. Karttunen has incorporated any of it into her "Analytical Dictionary". A bit like writing a lexicon of Golden Age Spanish and leaving Cervantes out. As well as the historians of the invasion of America.
There are many smaller difficulties. I am bothered by Ms. Karttunen's retention of Molina's habit of placing reflexive verb usage ("nino") first in the entry. Counterintuitive and awkward. And I would also point out that knowledge of Spanish is very helpful, as sometimes a Spanish translation is not carried over into an English gloss (the original Spanish entries are included).
A salient virtue of this book is the indication of vowel length and the saltillo, both phonemic in Nahuatl. However, Ms. Karttunen devotes what seems to me inordinate space in many entries to whether a given vowel is long or short, and who said what about the matter.
But this leads to the important point. In her Introduction, Ms. Karttunen says that one of the virtues of "computational linguistics" (meaning that the book derives from a computer database) is its appropriateness to a work in progress. This was back in the Seventies. It's high time that:
A) detailed discussion of vowel length etc. is replaced by
B) a thorough glossing of Sahagun and other important sources such as Chimalpahin.
This valuable work might be seen as the notes for a definitive dictionary of Classical Nahautl. It will be a very useful substitute until the new edition, based on this one, arrives.