Thanks Josh. I was actually aware of tori being the -masu stem of toru, but the books I checked did not mention that this stem could be used for nominalizing. (The primary reference I checked being Chino's "Japanese Verbs at a Glance.")
I guess my confusion with the -te form comes from the numerous examples like -te kudasai, -te aru, -te mo ii, -te oku, etc., where it is used very much like a gerund.
Ah, I see what you mean. The difference, I think, is between looking at the language as itself, versus decoding it into one's own language.
When I think "noun" with regards to Japanese, I think of a term that takes "da/desu", is put in the past aspect with "datta/deshita", and is negated by "ja/de wa nai". So, in that case, the "-te" form is not a noun (or in Japanese, "meishi"). On the other hand, words such as "kirei" (beautiful) and "onaji" (same) are nouns in Japan, even though the English translation marks them as adjectives.
Now, if one attempts to directly decode, say, "totte kudasai", you can get something like "give me taking" = "take (this)", and then you can refer to the "taking" as a gerund. But this is not how the Japanese look at it. As the "-te" form is conjunctive, it is connected to the next phrase. "Kudasai" is the imperative form of "kudasaru" -- literally meaning to send down, to give, to the speaker (or speaker's proxy). So the speaker is asking the listener to "take", and then adding the further modification of "kudasai" indicating in what manner the taking should be done. The same with "-te oku", "-te shimau", "-te mo ii". These phrases have developed idiomatic meanings which might have to be translated into the English gerund, but are really no different within the language from sentences such as "Kinou wa suupaa ni itte, sushi wo katta."