diagram this 94-word sentence

April 25, 2007 4:41 PM

How do I diagram this sentence?

Here is the sentence in question:

"The move from a structuralist account in which capital is understood to structure social relations in relatively homologous ways to a view of hegemony in which power relations are subject to repetition, convergence, and rearticulation brought the question of temporality into the thinking of structure, and marked a shift from a form of Althusserian theory that takes structural totalities as theoretical objects to one in which the insights into the contingent possibility of structure inaugurate a renewed conception of hegemony as bound up with the contingent sites and strategies of the rearticulation of power."

This sentence won the bad writing contest in 1998. Googling reveals several ways of diagramming sentences. I want to diagram it the way a contemporary linguist would, but I'm not sure exactly what that is.

posted by mai to writing & language (26 answers total)

Linguists generally don't use the old sentence diagrams, but I have seen many use a syntax tree to describe sentences.

Hand parsing this particular sentence wouldn't be too difficult, since it doesn't seem to have any really complex syntax, but it would take quite a while.
posted by demiurge at 5:21 PM on April 25, 2007

I'm not sure how you'd do this in HTML, but really, the heart of it is something like:

the move ---> brought

and

[the move] ---> marked a shift

That's a start.
posted by Robert Angelo at 5:43 PM on April 25, 2007

I'm not sure how to post a diagram here, but the basic structure is...

move [subject] | brought [verb] | question [dir obj]

both "move" and question are "modified" by the article "the", which would hang slant-wise off the modified words in the diagramming style I learned.

e.g.

move
t
 h
  e

move is also modified by a prepositional phrase: "from a structuralist account", which would hang like this

move
    f
     r
      o
       m
        account

and then "a" and "structuralist" would hang off account

I'm not 100% sure of this, but I think "in which" is a conjunction that links to a subordinate clause "capital is understood." This clause modifies "account" and the conjunction, "in which", links "account" and the clause. This means that "in which" should slant off the bottom of "account", but it ("in which") should have a dotted line under it. And the dotted line should hit the clause. (see these examples).

Okay, in the clause, we have...

capital [subj] | is understood to structure [verb phrase]

I think "to structure social relations" is the direct object of "is understood", and I suspect it follows a similar pattern to Sentence 52 here.

In any case, "in relatively homologous ways" is a prepositional phrase that hangs off "relations", and "to a view" is another prepositional phrase that hangs off "ways."

"In which..." is a (dotted line) conjunction that modified "view." It links to the clause "power relations are subject..."

relations [subj] | are [verb] | subject [dir object]

Hanging off "subject" is the prepositional phrase "to repetition, convergence and rearticulation." This is a three-way branch, similar to the and structures of 27, 29 and 30 here.

And that brings us back to...

move [subject] | brought [verb] | question [dir obj]

"of temporality" is a prepositional phrase that hangs off "question" (as does the article "the"), "into the thinking" is a prepositional that hangs off "temporality", and "of structure" is a prepositional phrase that hangs off "thinking."

"marked" is another verb liked to the original subject "move" by the conjunction "and"...

move [subject] | brought [verb] and moved [verb]

...so you'd need to diagram the two verbs as forked paths off of "move"

"a shift" is the direct object of "marked."

"from a form" is a prepositional phrase that hangs off "shift" and "of Althusserian theory" is a prepositional phrase that hangs off "form."

"takes structural totalities" is a clause that modifies "theory." It's linked to "theory" (via a dotted line) by "that".

takes [verb] | totalities [dir obj] (modified by structural [adj])

We then get another prepositional-phrase chain: "as theoretical objects" hangs off "totalities" and "to one" hangs off "objects,"

Once again, "in which" is a conjunction that links (dotted line) "one" to its modifying phrase the insights [sub] | inaugurate [verb] | conception [dir obj] (modified by "a" and "renewed").

"insights" is modified by the prepositional phrase "into the contingent possibility", and "possibility" is modified by the preposition phrase "of structure."

Jumping back to "the sights [sub] | inaugurate [verb] | conception [dir obj]", "conception" is modified by the prepositional phrase "of hegemony", which is modified by "as bound up" which is modified by "with the contingent sites and strategies" (fork at "and"), which is modified by "of the rearticulation", which is modified by "of power."

Whew!
posted by grumblebee at 6:50 PM on April 25, 2007

Well, I took a go at it.. Came up with a generic syntactic tree based on a few different versions of generative grammar all mashed up in my head..
posted by greatgefilte at 7:57 PM on April 25, 2007 [21 favorites]

Now, if you actually want a proper version of it, with binary branching and X-bar and everything, that'll take a little bit more than ten minutes and a sheet of scrap paper. :)
posted by greatgefilte at 8:02 PM on April 25, 2007

Great gefilte, Batman! That syntactic tree sure is a dandy!

To address the question of how to represent it online, when linguists need to show a strictly-branching parse without drawing a tree, they put square brackets around each constituent, with a little subscript by the opening bracket to state its category.

As a random example, here is 'capital is understood to structure social relations in relatively homologous ways', same parsing as greatgefilte's, but nongraphically represented:

[_S[_NP[_Ncapital]] [_Iis] [_VP[_Vunderstood] [_CP[_S[_Ito] [_VP[_Vstructure] [_NP[_AP[_Asocial]] [_Nrelations]] [_PP[_Pin] [_NP[_AP[_Degrelatively] [_Ahomologous]] [_Nways]]]]]]]]

Yes, that's a pain in the butt to read. It would be better if I used larger brackets for outer enclosing constituents, and smaller ones inside, as a responsible typographer would do. But there is sufficient hand-written tag vomit in this post as it is. Certain things ought to be done by a capable typesetting program, or not at all. Heck, it would also be easier to read if I pretty-printed it, with indentation and all, but that is something I have never seen an American structuralist do. (Might make things too accessible?)
posted by eritain at 2:52 AM on April 26, 2007

More modern grammar theories (such as those descended from GPSG such as HPSG) don't use trees but instead employ structures in which substructures may be shared. Here is one example of an extreme approach to axillary verbs for the sentence

John had eaten lunch

.
-ed(j=John, have(j, -en(j, eat(j, lunch))))

This says that the structure j is the subject of 4 different verb-like structures which compose the 2 words had and eaten.
posted by MonkeySaltedNuts at 7:51 AM on April 26, 2007

Thank you metafilter! Not only have you answered the question, but you have given me an opportunity to feel smug. My friend on behalf of whom I asked this question was skeptical that we would get any answers, but I knew the hive mind could do it.
posted by mai at 10:53 AM on April 26, 2007

Just to emphasize what demiurge said, this has nothing to do with linguists—sentence diagramming is a staple of old-school grammar classes, which were of course taught by English teachers, not linguists (and English teachers very rarely know the first thing about linguistics, which is why it's so depressing to know the first thing about linguistics and have to put up with what people say about language). Nothing wrong with diagramming—it's a pleasant and soothing activity, like knitting—it's just not something linguists do.
posted by languagehat at 11:02 AM on April 26, 2007 [2 favorites]

I don't see an essential difference between the old-style sentence diagrams and the various ways of representing parse trees — or DAGs in MonkeySaltedNuts's case — they're both, essentially, hierarchical decompositions of the sentence with a few symbols thrown in to annotate the meaning/interpretation of each subtree. I guess you can draw a sentence diagram without thinking in terms of a Chomsky-style grammar, and that's harder to do with a parse tree...
posted by hattifattener at 1:10 AM on April 27, 2007

Can I piggyback and ask MonkeySaltedNuts for a little illumination on phrase structure grammars? Something in the way of a ten-line overview? I am slightly acquainted with Link Grammar, so the idea of shared dependents isn't utterly thoughtcrimeful to me, but otherwise I've mostly been taught various flavors of Chomsky.
posted by eritain at 4:46 AM on April 27, 2007

ertain - 10 lines could never give an intelligible overview of any of the more modern theories. As a general term they are sometimes referred to as Unification Grammars. I was going to point you to a good online linguistics paper that uses Construction Grammar, but I found it has been removed. So, go to an academic library and find Kay, Paul and Charles J. Fillmore. 1999. Grammatical Constructions and Linguistic Generalizations. The What's X Doing Y? Construction. Language, 75(1): 1-33. (March 1999)

In a formal nutshell, most non UG theories consist of 1) structures and 2) rules that generate, transform, decorate, constrain, filter, or do something else to the structures. Some UG theories contain only structures (sometimes with inheritance) - the monotonic "unification" of which achieves everything done in other theories by phrase-structure rules, transformations, constraints, principles, redundancy rules, etc.

hattifattener: While some UG theories restrict themselves to DAGs there is no inherant reason a structure can't contain cycles and I've seen some examples that are much simpler that way (sorry but I don't have any references handy). With cycles you no longer have a hierarchy.
posted by MonkeySaltedNuts at 7:16 AM on April 27, 2007 [1 favorite]

Nothing wrong with diagramming—it's a pleasant and soothing activity, like knitting—it's just not something linguists do.
I loved it in junior high, and think it seriously helped us understand structure, and modifiers, and order, etc. We hadn't ever really thought about the bones of it all before. (Also, most of us were starting a foreign language then too, so it helped with that in an oblique way, since the basic structures are different.)
posted by amberglow at 8:40 AM on April 27, 2007

Do kids still diagram in school, or has whole-language, etc, killed it off?
posted by amberglow at 8:41 AM on April 27, 2007

You guys want me to put the tree on a t-shirt or something?
posted by greatgefilte at 11:12 AM on April 27, 2007

So to paraphrase: "structuralism suggested that society was structured by capital, and that the relationship of capital to structure was static. Then some people decided that you should look at the churning of power, which is dynamic." Right?
posted by honest knave at 11:34 AM on April 27, 2007

greatgefilte, how about a temporary stick-on tattoo? :)
posted by junesix at 4:18 PM on April 27, 2007

Even after Honest Knave's paraphrase, I still don't understand what the sentence was trying to say. I'm getting a cold sweat and bad trip memories of junior high school and an english teacher who thought diagramming sentences would someday cure cancer, bring about world peace, and make her young again.
posted by ZachsMind at 6:55 PM on April 27, 2007

I get down on my knees and thank god every single day that I have never, thus far, spent 1 precious millisecond of my existence learning or practicing the art of diagramming a sentence. Now, as I daily bend my knee to express my gratitude to the lord for this small miracle, I shall also give thanks that I have never been impelled to read whatever article or book that ghastly, nay, horrific sentence came from. Thank you metafilter, for teaching me to appreciate my life.
posted by serazin at 9:41 PM on April 27, 2007 [2 favorites]

That won a bad sentence contest? That sentence is great! It sounds like Lovecraft covering Marx.
posted by BlackLeotardFront at 9:56 AM on April 28, 2007 [2 favorites]

As a onetime Comp. Lit. minor, I'll try to help ZachsMind and put this in English.

"Originally, the structuralist movement (i.e. early 20th-century Marxist critics, particularly Althusser) argued that social relationships were governed by power expressed in the form of capital. The more capital you had, the more power you had. As a theory this was found to be too rigid and did not express how power relationships changed over time. Critics in the post-structuralist period began to look at how social relationships hinged particularly on how strengths and weaknesses of power were changing, bringing a more dynamic understanding of how these social relationships can be analyzed."

It's not really a bad sentence, it just contains a passel of post-structuralist jargon that can throw you if you don't understand how the terms have been appropriated for use in that theoretical system.

To be honest, most post-structuralist writing exploits terminology in ways that make an average sentence about these concepts opaque. I could go into how this is a broad philosophical pun, in that the writing itself articulates power in ways that are all but impossible to understand without a full, exhaustive grounding in the discipline, which is almost impossible to get without buying into the political basis of it all, thus writings about power are demonstrating the very ideas they describe. But most people wouldn't care ...
posted by dhartung at 4:07 PM on April 28, 2007 [7 favorites]

The Bad Writing Contest where this came from, for more context.
posted by dhartung at 4:19 PM on April 28, 2007

dhartung: awesome. And knowing that Judith Butler wrote that fills me with unspeakable happiness.
posted by carmen at 9:09 PM on April 28, 2007

Whole-language etc. has done a number on the teachers. The best (and nearly the only) grammar teacher I had growing up was just barely able to teach us diagramming. The second-best could identify gerunds and participles, but could not clearly explain the difference; likewise she could identify passive voice, and had been properly inculcated with the terror thereof, but she couldn't explain it well enough to tell us what it was when she said not to use it. So instead she told us to avoid be entirely, and I wrote in e-prime for a year.

Actually, I snuck two copulas past her, when it would have been really, really barbarous to do otherwise. And then I took two years of Latin and got some idea what was what, just in time for the Latin teacher to retire and nobody else in the district to be competent to teach it. So yeah, I'd say the ability to explicitly discuss the grammar of a language is mostly no longer conveyed in U.S. schools.
posted by eritain at 10:06 PM on April 29, 2007

Thanks, MonkeySaltedNuts—third paragraph of your first link is exactly what I was looking for.
posted by eritain at 4:28 AM on April 30, 2007

I love you, hive mind.
posted by DenOfSizer at 8:34 AM on May 6, 2007