Linguistic commentary from a guy who takes things too literally

Lincoln’s Subordinate RNR

Posted by Neal on February 12, 2009

lincolnvc26p7I never realized that Abraham Lincoln and Charles Darwin shared a birthdate, but the fact has been inescapable for the past month or so, what with it being the 200th anniversary of that date, Feb. 12, 1809. The spotlight on Lincoln reminds me of something I noticed a month or so ago, when Doug wanted to learn about the (American) Civil War. I pulled out our three-pound, foot-wide set of the 1990 Ken Burns miniseries on VHS. (My birthday present from 1998.) On the first volume, as we listened to Sam Waterston reading parts of Lincoln’s first inaugural address aloud, this line caught my ear:

We are not enemies, but friends. We must not be enemies. Though passion may have strained it must not break our bonds of affection.

Something was unusual about the last line. Two partial clauses, passion may have strained and it must not break, with a shared direct object, our bonds of affection. At first, I thought that last line contained another parasitic gap, like the one I wrote about a couple of posts ago. But then I realized that this sentence didn’t pass one of the tests for parasitic gaps. Was there any kind of filler-gap dependency going on, like a wh-question or a topicalization or a relative clause? No. The direct object came right after the second verb, just where you’d expect it after a pair of coordinated transitive verbs, as in Adam likes, but Doug dislikes, peanut-butter-and-banana sandwiches (a construction sometimes known as right node raising). There isn’t a gap where the direct object should be; there’s an actual direct object!

So then I wondered what had made me think Lincoln’s sentence was unusual at all. If Adam likes, but Doug dislikes, peanut-butter-and-banana sandwiches was nothing remarkable, with two partial clauses followed by one direct object, what was so special about Though passion may have strained it must not break our bonds of affection?

I finally realized: the partial clauses aren’t coordinated. There’s not an and, but, or or to be found in the sentence. What joins the partial clauses is though, a subordinating conjunction. Whereas coordinating conjunctions in English always come between the (last) two items they’re coordinating, subordinating conjunctions don’t. Look at the difference:

Passion may have strained, but it must not break the bonds of our affection.
*But passion must not break, it may have strained the bonds of our affection.
*But passion may have strained, it must not break the bonds of our affection.
Passion must not break, though it may have strained the bonds of our affection.
Though passion may have strained, it must not break the bonds of our affection.

I always thought right node raising was associated strictly with coordination. That was why this double-gapped case of subordination had sent me down a trail chasing after parasitic gaps. Looking in CGEL, however, I see that RNR-like subordinate structures are not unheard-of:

Delayed right constituents occur predoninantly in coordination, but they are found in some subordinative constructions too:
[27] i. I enjoyed, although everyone else seemed to find fault with, her new novel.
ii. Those who voted against far outnumbered those who voted for my father’s motion. (p. 1344)

Sentences like these are troublesome for syntactic analyses. If you know of an analysis that covers this RNR-like construction in subordinate structures, let me know in the comments!

add to : Bookmark Post in Technorati : Add to Blinkslist : add to furl : Digg it : add to ma.gnolia : Stumble It! : add to simpy : seed the vine : : : post to facebook : Bookmark on Google

2 Responses to “Lincoln’s Subordinate RNR”

  1. Hi,
    I’d like to plug my own paper, if I may. In my 2001 paper “The syntax and semantics of left-node raising in Japanese” (which is available at, I propose an analysis that can handle right-node raising (and what I call left-node raising) involving non-coordinate structure. Rui Chaves’ 2007 dissertation, available at, also discusses this phenomenon. These are both HPSG-based analyses, but it’s not too difficult to develop a Categorial Grammar-based account of this phenomenon, either, is it?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: