Colorless Green Ideas Sleep Furiously Blogging

This is well-worth tracking down.

Hoisted from comments:

Jonathan Goldberg: Talk of colorless green ideas is not necessarily incoherent. This was proven to me by a poem that served as the intro page to a volume of essays (to which, alas, I've lost the reference) on Philosophy of Language.

It was three or four verses starting with the conceit of a seed in winter as the idea of the tree it would become in the spring. The last line, of course, was "Colorless green ideas sleep furiously." By the time one got there it made perfect sense. Brilliant.

Comments

[Jason]

Several examples, apparently from a 1985 Stanford competition, can be found in the middle of http://www.umich.edu/~archive/linguistics/linguist.list/volume.2/no.451-500

Jonathan Goldberg is probably thinking of one of these two:

It can only be the thought of verdure to come, which prompts us in the autumn to buy these dormant white lumps of vegetable matter covered by a brown papery skin, and lovingly to plant them and care for them. It is a marvel to me that under this cover they are labouring unseen at such a rate within to give us the sudden awesome beauty of spring flowering bulbs. While winter reigns the earth reposes but these colourless green ideas sleep furiously.

C. M. Street

Behold the pent-up power of the winter tree;
Leafless it stands, in lifeless slumber.
Yet its very resting is revival and renewal:
Inside the dark gnarled world of trunk and roots,
Cradled in the chemistry of cell and sap,
Colourless green ideas sleep furiously
In deep and dedicated doormancy,
Concentrating, conserving, constructing:
Knowing, by some ancient quantum law
Of chlorophyll and sun
That come the sudden surge of spring,
Dreams become reality, and ideas action.

Bryan O. Wright

[Curtis Brown]

I'm pretty sure the poem in question is this one:

John Hollander

Coiled Alizarine
for Noam Chomsky

Curiously deep, the slumber of crimson thoughts:
While breathless, in stodgy viridian
Colorless green ideas sleep furiously.

from The Night Mirror, by John Hollander. ©1971

(I'm pretty sure it's this one because I also remember seeing it at the beginning of a philosophy of language collection -- it might have been the collection of essays on Chomsky edited by Gilbert Harman, but I don't have my books handy right now to check.)

The poem can be found here: http://web.mit.edu/24.900/www/poem.html.

[Bob Violence]

"Colorless green ideas sleep furiously" was originally composed by Noam Chomsky as an example of a sentence that is perfectly grammatical, but doesn't make sense. There are several poems, not just from the competition mentioned by Jason, that use the sentence.

http://en.wikipedia.org/wiki/Colorless_green_ideas_sleep_furiously

[Jon Fernquest]

Lakoff's book "Beyond Cool Metaphor" discusses poetic metaphor in detail, including the sort of all consuming poetic conceits one finds in John Donne's poetry, but what it doesn't cover is the seemingly nonsensical surrealist metaphor like Lautreamont's famous: "Beautiful as the chance encounter of a sewing machine and an umbrella on a dissection table" the same sort of nonsensical but grammatical sentence as Chomsky's "Colorless Green Ideas Sleep Furiously Blogging"

The beauty and allure of surrealist metaphor though is that the sentences usually have some ways in which they make sense, provoking the reader to make further sense out of them. One of these days work by computational lexicographers such as James Pustejovsky (Brandeis University) may help reveal how this works. I'd recommend French poets like Benjamin Peret or Andre Breton. Dadaists like Tzara often violate grammatical constraints.

[Scott Martens]

This is actually an important point in linguistics. Chomsky dwelled a lot on the idea that a sentence could be well formed yet meaningless to distinguish between linguistic competence as an abstract application of syntactic rules and the dirty business of actually communicating. But it's not clear that any sentence - even non well formed ones - has to be meaningless at all given an appropriate context. This undermines the notion that syntax can be studied in the abstract, free from discussion of the real conditions of communication. It blurs the divide between performance and competence, and since performance does not lend itself to neat mathematical models very well, tends to undermine claims about universal syntactic properties of languages.

Although Pereira's result using statistical models of bigrams - showing that quite basic statistical models can distinguish between novel well formed and non well formed sentences - is a much more direct attack on the notion.

[kid bitzer]

if these are being presented as counter-evidence to the charge of incoherence, then they fail.

All that they show is that the a lot of poetry is incoherent, which we knew independently.

But for the sentence really to be coherent, it would have to have a model, i.e. a set of objects that verify it. And it doesn't.

Because if you really try to make sure that "ideas" means "ideas" (rather than a seed), then the word "sleep" can no longer be truely predicated of the subject.

And if you really mean by "colourless" that the object to which that property is attributed has no colour, then the property of greenness no longer applies to it. Sure, maybe some other property does (potentially-green, e.g.). But not "green".

And so on.

The sentence is still incoherent--there is no stable assignment of literal senses that can make it turn out true.

Of course, that's not how we read a poem--in that kind of activity, we are happy to let the same thing be "colourless" in one non-literal sense and "green" in some other non-literal sense that does not conflict with it. We're happy to use "sleep" as a metaphor for a pre-germinant phase of a seed. We're happy to say that the seed is somehow an idea, even though no seed is an idea and no idea is a seed.

So, I mean, bravo, great stuff poetry and all. Wonderful, evocative, polysemic, neato.

But if you actually try to construct a model for it, you fail.

And this is actually an important point in linguistics, too--poetry cannot demonstrate coherence, because poetic usage does not demand coherence (and frequently flouts it).

[RW]

Focusing on the communicative aspects of language and the relevant requirements of syntax would appear to unduly constrain the problem of language for two fundamental reasons, one of which the poem illustrates quite well: No accomplished English speaker would have difficulty discerning the meaning of "colorless green ideas sleep furiously" within the context of the poem and, even in isolation, would probably find the phrase at least as meaningful (and possibly more intriguing) as "pass the salt, Walt."

Secondly, as George Steiner* wisely proposed (he's a 'mere translator' rather than a linguist and in consequence is probably rarely cited in the academic literature) language has two fundamental functions other than communication: (a) invention -- by extension not oriented to veracity -- and (b) group identification; that is, from an evolutionary perspective, the dominant features of human language are creative and hermetic rather than informational.

Language alone could not free humanity, certainly a strongly syntax bound language could not; that project required the invention of metaphor, tall tales (all right, fibbing) and, quite crucially, the future tense (what else is imagination if it is not the ability to say something as if it were already so or might become).

*Steiner, George (1992). After Babel: Aspects of language and translation. Oxford University Press. 2nd Ed.

[J.Goodwin]

If anything, Chomsky's mistake in forming his example was assigning properties to something that is by definition intangible. Intangible items are shaded (in a manner of speaking), chaotic, stateless and can easily exude properties that in a more tangible sense would be, at least in some sense, mutually exclusive.

In a way, he seems to have subconsciously derived a sentence that makes itself sensible and potentially meaningful. And then readers have themselves sought to find the meaning of the sentence, and create sense of it. I find that comforting, actually. It also speaks to the powers of the human mind.

[J.Goodwin]

It's possibly also worth pointing out that the vast majority of actual verbal communication is "incoherent," as compared with written forms of language. We very rarely adhere strictly to the "rules" of our languages.

[ferd]

I know there's something real going on here.

But, I wonder if any academic ever made a living spouting brilliantly unintelligible nonsense his or her entire career.

[WatchfulBabbler]

Did any academic *not*?

[Richard S]

Thirty years ago I recall having the same discussion topic in a philosophy of language class. I thought then, and still think that Shakespeare had preempted, and unmade Chomsky's argument with these quotations from Hamlet:

Though yet of Hamlet our dear brother's death
The memory be green, and that it us befitted
To bear our hearts in grief and our whole kingdom
To be contracted in one brow of woe,
Claudius: Hamlet Act I, scene II

Affection! pooh! you speak like a green girl,
Unsifted in such perilous circumstance.
Polonius: Hamlet Act I, scene III

First, her father slain:
Next, your son gone; and he most violent author
Of his own just remove: the people muddied,
Thick and unwholesome in their thoughts and whispers,
For good Polonius' death; and we have done but greenly,
In hugger-mugger to inter him:
Claudius: Hamlet Act IV, scene V

[Paul Reber]

Brad, I strongly suspect, is like me, at least partly captured by the curiousity of poetry aimed at Chomsky's famous sentence because he may have noticed the subtle riff on this idea included by Charlie Stross in his novel Accelerando which he has lauded here (which was also the original reason I started reading this blog, oddly enough).

It was dropped as an aside amongst the description of the enveloping singularity in intelligence, if I recall correctly, without explanation or further discussion. Stross has many such subtle throwaways in that novel and reading it I am compelled to wonder how many other subtle points are hiding in there that I didn't understand. I had never emailed an author prior to reading his stories from Accelerando but I was moved by the obvious effort Stross put into getting details right from a Cognitive Science, Neuroscience and Computer Science perspective that I simply had to congratulate him. I can only wonder if he got the majority of his Economics riffs correct, too.

I suppose this is going to look like a thread hijack. Or perhaps you can just consider my reply to be poety.

On the topic of sense & syntax, I am not a linguist (although there are a fair number not far from my academic office), it is my impression that if there is effective communication, then the sentence is fine by me.

[Mandos]

Although Pereira's result using statistical models of bigrams - showing that quite basic statistical models can distinguish between novel well formed and non well formed sentences - is a much more direct attack on the notion.

So like, I don't think it shows much more than if you have enough events, they'll fit some kind of distribution. And then you just futz around with the training algorithm until you get the one you want.

I mean, it's a 78 million-word training corpus they used or something, if you're talking about Saul and Pereira 97.

Wake me up when they have a theory for sluicing and stuff like that.

[Mandos]

D'oh! Forgot that this blog eats italics tags. The first paragraph belongs to Scott Martens.

[frenchman]

The point of Chomsky's made-up example sentence was simply this: the impression we have, as native speakers, of what constitutes a complete sentence (an utterance) tells us that "Colorless green ideas sleep furiously" is one. It's the laws of semantics that are "violated" by the juxtaposition of two adjectives, colorless and green, that are logically incompatible. Therefore, they cannot modify their noun in the usual manner.

Chomsky used to say, and wrote, that he came on the basic concept of linguistic competence while doing a research project in linguistics. He was supposed to derive a grammar of Hebrew by talking to "informants" like Boaz did. But he already knew Hebrew. He could be his own informant. He realized then that the capacity to produce sentences in a language was a basic skill that could be exercised in many different ways. In fact, it was so basic that it governed verbal interaction even in cases where there were non-standard and non-agreed on meanings for words. To put it another way: the sentence was the vehicle of meaning. We've all had cars that could get you there despite many problems under the hood. Sentences are like that. Our ability to tell when they begin and end allows us to run their possible meanings through our minds in an instant, and we "construe" whatever we don't understand.

To think of "colorless green" as anything but an anomaly takes us into the world of poetry - sort of. Chomsky certainly wouldn't care, past that point.

Now, the beat downs.

Goldberg: Try maybe "The force that through the green fuse drives the flower..." A little less Spencerian (retorts!), and it allows us to avoid the ugly word, "conceit". Words are contagious in this realm, Mr. Goldberg. One cannot fashion beauty from ugliness. And "winter reigning"? I hate when that happens, I get all weght. (Not to mention that Dylan T. is an actual poet. You can't just call yourself a poet, you know. That is one word that doesn't just stick.)

Contestant #2, be it Street or Wright, is at least a dedicated doorman (unless doormancy has to do with necromancy, or something). Only the pocket-protected could come up with "Cradled in chemistry". By an ancient quantum law, this turns out to generate no meaning at all (not in terms of poetry, anyway). Also, I love the way the last line really hammers home the INESCAPABLE CONCLUSION: we are like trees. Or maybe the trees are getting ready to do something they'd always dreamed of, like falling over on this fucksnot of a mother's tongue.

#3: The New Yorker rule of poems is: always use one unusual word. Here you have two, counting alizarine, a synonym for crimson (cp. Donovan's "Wear Your Love Like Heaven") which probably came out the same year). Viridian means green. John Hollander wrote books explaining poetry to high school students, so that tells you which Circle of Hell he inhabits. If a little New Yorker archness is good, is a little more better? The deconstruction of his repulsive little jape is left for you.

#4: Jon Fernquest has been killed and eaten by Andre Breton in the form of a cockroach. Lakoff (pronounced like back off) is someone who thinks of words as little densepack explosions happening in your head, each giving you just that little bit of information.

Lakoff is famous for talking about memes and framing, but that's with regard to commercial speech. Memes are phrases that tend to get repeated and to insert themselves into conversation (for a time). Frames are actually our prejudices and unexamined assumptions. Lakoff clearly assumes, without ever thinking he has to argue this part of it, that people are mentally engineered like this. Frequent repetitions of memes and re-presentations of frames in statistical terms generate movement in the electorate (or in shoppers).

This means Lakoff is a behaviorist. Nothing unusual about that: all Americans are behaviorists, except maybe Chomsky and the hundreds of millions who would go "Wha?" if you tried to tell them their minds work like this. No. Their purchases work like this - sometimes - maybe.

But to say poetry works like this is simply to annul the category. Lakoff cannot tell the difference between a good poem and a bad one. His definition of a poem would have to include the laundry list from Foucault's Pendulum, and a lot of things like that. The father of behaviorism, Watson, was said to dream without images. he lacked the power to visualize, and suspected all his life that people who talked about dream imagery were just making things up. Blind, he refused to admit the existence of sight. Lakoff, deaf, refuses to admit the existence of sound, and that's why he gets adopted by the Sunday-puzzle mavens of computational linguistics. Computers sleep furiously but quietly. Just keep putting the great texts into searchable databases, boys. We'll handle the heavy reading. A mind is a terrible thing to paste. And what did you think you were going to do, anyway, after your computer analysis "revealed" how poetic metaphor "works"? Understand romance? Get dates?

#5: Note to Scott Martens: "syntax can be studied". Don't tell me, Scott. Show me. Do it.

#6: Dear Kid Bitzer. If you want to say that poetry is incoherent, it behooves you to be coherent. You do not know "independently" that poetry is incoherent. You know this idiosyncratically, that is, you have learned to say this to mask off the painful truth that the coherence of poetry - attested by the wise and great throughout the centuries, and by John Hollander - is something you can't grasp. And this says something about you, kid. There's a whole bunch of neat stuff about poetry you are left out of, because you can't construct a model of it.

But we don't construct models, Kid. Poets construct singular objects, some falling into genres, and some not. No one thinks you can learn to write poetry, assuming that was the point of your impossible model.

The opposite of running a model is something you may not have considered, Kid. It is singular, unrepeatable human action that has moral bearing beyond the norm, such that the category of beauty is invoked to describe the coherence of that action with the armature of Goodness: "Beauty is the Symbol of the Morally Good" (Krit. d. Urteils. section 59).

Or in words you may find coherent: poets actually do something - but you're not allowed to do it. You aren't able to do it. All you can do is talk about it.

Get it? You're flouted.

RW: the laborers of the Tower were not speaking poetry. They were just ordinary people who suddenly couldn't understand a fucking thing. Tell me - do you ever dream there's tape over your mouth?

J. Goodwin - can I come along, sometime, when you speak to the powers of the human mind? I may know some of them.

Paul Reber - are you Lakoffian? I have long wondered just exactly how many academics cherish the illusion that just because there are a bunch of supposedly smart people in geographical proximity, it's going to rub off. No need to worry about hijacking the thread. The passengers have subdued you.

The point should have been that you can create a context for a sentence that is non-sensical in common-sensical terms. Chomsky was just making up an example sentence, only for use in an argument. He probably did have some conundrum about ideas, consciousness, and in-der-Brain-sein running through his Lakoffian cerebellum. But the fact that it sort of "took off" never really managed to have anything to do with real poetry. For that, you'd have to ask what "colorless green" looks like. Or what could be colorless and green.

Or you would have to be lucky enough to see it, like this guy, who got out of his car on the Interstate in Iowa to watch a storm sweep over a cornfield.

I’m stopped on the dim ramp. Semis whine
And thump the buckles of the overpass.
Furious my irises gather
By force light in the interchange
To keep me focused on the page.
Dark’s deeper here. Mare-licked,
The starlight’s lather – splashed and low
Clouds ferry light from Iowa City.
Corn before storm sleeps (colorless green sleeps furiously)
And a corporal’s guard of lightning bugs
(all I brought with me?) regroups to face rain.

(Gerry Strachan)

[Mandos]

frenchman, you are my hero.

[Scott Martens]

Frenchman, show what exactly? My whole PhD thesis is the development of an approach that can, in opposition to Chomsky, extract highly detailed and hopefully very comprehensive syntactic and morphological rules from texts themselves. But I'm hardly the first person to make such an effort - the literature is full of approaches of that type, mostly at least somewhat successful.

"This means Lakoff is a behaviorist" - surely you jest. The core of the behaviorist methodology was to treat internal mental states as outside the scope of scientific study. Lakoff is positively obsessed with connectionist models of the kind of behaviour he describes. It may be a bit silly - IMO it is - but this is hardly treating the brain as a black box.

Mandos, so? Peirrera found that a quite simple bigram model could do empirically what Chomsky's "colorless green ideas" example was meant to show couldn't be done: distinguish between well-formed and non-well-formed sentences. Chomsky's entire introspective method rests on the claim that an empirical linguistics must necessarily be inadequate. This claim is pretty weak nowadays.

[Jon Fernquest]

"Jon Fernquest has been killed and eaten by Andre Breton in the form of a cockroach."

I made sure that didn't happen.
When I lived in Yangon.
Quantity, not size, was the problem.

Frenchman might not be real either, one can never tell in the cyberworld. I could imagine a program that negated whatever one said, added a dash of perversity, and echoed it back. One might even consume oodles of calories interacting with it, only to realize that it wasn't human.

Frenchman seems to be saying that any attempt to reduce part of human language to computation is necessarily behaviourist.

Computation can also be used to **augment** human writing, synthesizers are used to compose symphonies, a thesaurus that generates whole phrases with tropes, rhetorical, figurative, arrayed on a pallette running down the side of the screen, combinatorially beyond what a thesaurus can do nowadays, tailoring itself to one's writing habits like a Google sidebar.

"Lakoff cannot tell the difference between a good poem and a bad one"

If the language generated computationally was indistinguishable from human language, in a Turing test let's say, a charge of "behaviouralism" wouldn't have much meaning (Skinner is quite a frightening intellectual bogeyman that one must quite quickly disassociate oneself from, agreed)

[Mandos]

"Frenchman, show what exactly? My whole PhD thesis is the development of an approach that can, in opposition to Chomsky, extract highly detailed and hopefully very comprehensive syntactic and morphological rules from texts themselves."

That's not the point. You can find all kinds of patterns by looking at things in all kinds of ways. As Perreira notes in a later review paper, machine learning has given us some interesting tools to characterize the size of the hypothesis space and even structure previously intractable spaces to make them seem more tractable using clever mathematical tricks. This is especially true given our thankfully much better computer power.

But this tells us nothing about what the *right* hypothesis-generating machine is for characterizing the human mind and the phenomena we observe in language.

Lord knows grammar induction is a hot topic, and it may even be a *useful* topic, *but it is not a science of language*. Being able to generalize from a corpus doesn't mean that you have the *right* generalization from the corpus that corresponds to the human language machinery. So you might (I don't know) be able to identify a phenomenon such as sluicing, but that's not the point: what is the *principle* that unifies these phenomena?

If you think you can short-circuit this directly via neuroscience or some sort of Pinkerian theory of language evolution or something, well...but you ain't short-circuiting it via grammar induction, nope.

"Peirrera found that a quite simple bigram model could do empirically what Chomsky's "colorless green ideas" example was meant to show couldn't be done: distinguish between well-formed and non-well-formed sentences."

Under extremely unrealistic conditions, is the point. Yes, with our increased computer processing power since the 50s, giant hard drive space, etc, you can crunch the numbers and simulate, individually, a number of behaviours once considered human-specific. Huzzah.

As Chomsky himself puts it these days, you can do all sorts of things with a bigger bulldozer. I mean, we've managed to simulate flight for humans quite nicely now---with birds that don't flap their wings (aka airplanes). It's useful, but it ain't a science of bird flight.

Chomsky's idea will be a bit more challenged when you can put a computer in a realistically human developmental setting and have it make the syntactic generalizations that one of those linguistic idiot savant cases make. Otherwise...you've got a nice pattern recognizer. A very good, perhaps very useful tool. There are people who I think are on the right track for this; for instance, I understand that Janet Fodor and Bill Sakas, simulate learning by constraining the grammatical hypothesis space with a lattice of...parameter sets.

As far as I can tell, computational linguistics (being sort of in the business myself as you could probably tell) have developed this conceit because they've lately been *very weakly* successful at performing certain highly artificial linguistic and information retrieval tasks using relatively linguistically-naive techniques. From this, some of them seem to believe they can resurrect some kind of more sophisticated behaviourism with clever smoothing magic and HMM-like doodads and all that good stuff.

I think there's a place for computational linguistics in all of this. It's just not that one. It's probably more along the lines of the recent work of John Hale at MSU and other such things.

[anne]

"What a fine post and thread," she said, in proper behaviorist style.

[Paul Reber]

The point of language is communication and therefore obsessing over the structure of syntax irrespective of context seems to me to miss an important point.

For example, when Stross dropped "colorless green ideas slept furiously" in his novel, I didn't worry about grammar, I took the meaning of the sentence as "I have read something of Chomsky and the question of language, structure and meaning." And that communication added value to my enjoyment of the writing.

As another example, when I said I was in close proximity of linguists, I believe I communicated that I am currently in an academic environment (which may not have been otherwise obvious). I was also attempting to mimic (perhaps poorly to overly subtly) Brad's occasional reference to a sense of being surrounded by many clever people in a rich University environment. In fact, feeling that there are many linguists around me, I feel somewhat chagrined that I haven't had opportunities to discuss the ideas embedded in Chomsky's example with actual experts. I suspect many academics feel (as I do) a semi-constant sense of disappointment that more of the good ideas of my clever colleagues are actually unable to rub off on my as I simply cannot find the time to spend more of it interacting with them.

See, context is important in communication. Last example: today I heard and then saw a woodpecker pecking away at a streetlight. As a scientist, I was first amused, then saddened then made anxious by this encounter.

And now back to work on today's research project, which I fervently hope is a tree and not a streetlight...

[Mandos]

Language has a "point"? You mean, in a teleological sense?

[anne]

Paul Reber:

"[T]oday I heard and then saw a woodpecker pecking away at a streetlight. As a scientist, I was first amused, then saddened then made anxious by this encounter."

Please describe further or attempt an explanation, if this is not an exercise. There may well have been insects along the post or on the glass. Still....

[anne]

"Talk of colorless green ideas is not necessarily incoherent."

No; there is a charm to the words as such, even to the word pairs, but they are incoherent as set down in the sentence. Such talk is necessarily incoherent, apart from the proper syntax.

[anne]

Frenchman:

'The point of Chomsky's made-up example sentence was simply this: the impression we have, as native speakers, of what constitutes a complete sentence (an utterance) tells us that "Colorless green ideas sleep furiously" is one. It's the laws of semantics that are "violated" by the juxtaposition of two adjectives, colorless and green, that are logically incompatible. Therefore, they cannot modify their noun in the usual manner.

'Chomsky used to say, and wrote, that he came on the basic concept of linguistic competence while doing a research project in linguistics. He was supposed to derive a grammar of Hebrew by talking to "informants" like Boaz did. But he already knew Hebrew. He could be his own informant. He realized then that the capacity to produce sentences in a language was a basic skill that could be exercised in many different ways. In fact, it was so basic that it governed verbal interaction even in cases where there were non-standard and non-agreed on meanings for words. To put it another way: the sentence was the vehicle of meaning. We've all had cars that could get you there despite many problems under the hood. Sentences are like that. Our ability to tell when they begin and end allows us to run their possible meanings through our minds in an instant, and we "construe" whatever we don't understand.

'To think of "colorless green" as anything but an anomaly takes us into the world of poetry - sort of. Chomsky certainly wouldn't care, past that point.'

Perfectly explained.

[Paul Reber]

Paul Reber:

"[T]oday I heard and then saw a woodpecker pecking away at a streetlight. As a scientist, I was first amused, then saddened then made anxious by this encounter."

Please describe further or attempt an explanation, if this is not an exercise. There may well have been insects along the post or on the glass. Still....
-------------

I'm not normally a bird watcher, but I heard the rat-a-tat-tat while walking the dog this am and realized there was no tree on the corner where I stood. The woodpecker appeared to be hammering away at the flat, opaque plastic that appears to replace glass in our neighborhood street lights.

I suppose there could have been bugs, or even that the woodpecker was making progress delving into the plastic (although it didn't sound like it). However, it looked to me like the woodpecker was attempting futiley to dig into an impervious surface having incorrectly mistaken the street lamp for a tree. If the bird had been pecking insects off the surface of the plastice, I would not have expected to hear such insistent hammering.

But, as I said I'm not a bird watcher (nor linguist, nor economist). After initial amusement at the inconguity, then slight sadness to see the bird stuck on what appeared to be a pretty bad strategy, a distant analogy came to mind: sometimes I attempt to "hammer away" at scientific questions that I hope will yield nuggets of intellectual sustenance. However on some days, it feels more like hammering at a lamp post than a nice tree full of yummy bugs.

[anne]

Paul Reber, than you so much for the details. I have no explanation, and whether this is worrisome or mildly sad, so I will ask a friend who knows of woodpeckers. The sense I have is that I have heard a woodpecker working away at a metal drain pipe, but I did not think carefully about why; I wonder if I really did hear such a sound? Again, I will think and ask.

[anne]

http://www.calvorn.com/gallery/photo.php?photo=6293&exhibition=7&u=99|18|...

Downy Woodpecker and House Sparrows Having a Dispute
New York City--Central Park, The Ramble.

[anne]

http://www.calvorn.com/gallery/photo.php?photo=6052&u=4|5|...

Downy Woodpecker and House Sparrow Converse
New York City--Central Park, The Ramble.

[anne]

http://www.calvorn.com/gallery/photo.php?photo=5361

Red-bellied Woodpecker
New York City--Central Park, The Ravine.