Is it possible to do a concordance using multi word phrases rather than single words?

[lucian303]

I'm wondering if it's possible to do something like the below, adopted from the concordancing documentation page, to search for multi word phrases rather than single words, and if so, how? I wrote the pseudo code that hopefully shows what I'm asking more clearly. Or does buzz strictly work at the token (word and punctuation) level?

ing = corpus.just.phrases('software architecture') # <-- modified to pseudo code since phrases() doesn't exist
ing.conc()