In the course of time, the sounds of a language depend on the gradual change of its articulation. We can pronounce a sound only when our memory retains an imprint of its articulation for us. If all our articulations of a given sound were reflected in this imprint in equal measure, and if the imprint represented an average of all these articulations, we, with this guidance, would always perform the articulation in question approximately the same way. But the most recent (in time) articulations, together with their fortuitous deviations, are retained by the memory far more forcefully than the earlier ones. Thus, neglibible deviations acquire the capacity to grow progressively greater…

Silverman goes on to mention some of Kruszewski’s other major insights, such as: (1) the arbitrary relationship between sound and meaning, (2) the non-teleological nature of the linguistic system, (3) the generative or creative character of language, (4) the connectionist organisation of the lexicon, and (5) the optimality-theoretic-esque proposal that linguistic systems may be analysed as the product of pressures and constraints in inherent conflict with one another. There is also a brief mention of Darwin’s influence on Kruszewski’s work (as we can see in his non-teleological stance).

The story ends on somewhat of a sad note, with Kruszweski suffering from a debilitating neurological and mental illness that cut short his promising career at the age of 36 — making his depth of scholarship and theoretical insight all the more impressive given it was produced in just eight years.

Anyway, you should take a look at the two articles, if only for an historical perspective on linguistics, but I would also suggest having a gander at some of Silverman’s other papers. He’s got his own ideas and insights that are worth considering (if you can wait, I’ll be discussing some of these in one of my next posts).

(1) embeddedness of phrases within other phrases, which entails keeping track of long-distance dependencies among phrases;

(2) the specification of the computed output string itself, including meta-recursion, where recursion is both the recipe for an utterance and the overarching process that creates and executes the recipe.

The case of grammatical recursion (see definition 1) is perhaps most famously associated with Noam Chomsky. Not only does he claim all human languages are recursive, but also that this ability is biologically hardwired as part of our genetic makeup. Countering Chomsky’s first claim is the debate surrounding a small Amazonian tribe called the Pirahã: even though they show signs of recursion, such as the ability to recursively embed structures within stories, the Pirahã grammar is claimed not to recursively embed phrases within other phrases. If true, then are numerous implications for a wide variety of fields in linguistics, but this is still an unsubstantiated claim: for the most part, we are relying on one specific researcher (Daniel Everett) who, despite having dedicated a large portion of his life to studying the tribe, could very well have been misled. That said, I retain a large amount of respect for Everett, having watched him speak at Edinburgh a few years ago and read his book on the topic: Don’t Sleep, There are Snakes: Life and Language in the Amazonian Jungle.

So, why am I rambling on about recursion? Well, besides its obvious relevance, — and perhaps under-representation on this blog (deserved or not, I’ll let you decide) — Everrett has recently published a series of slides about a corpus study of Pirahã grammar (see below).

Download (PDF, 841.58KB)

His tentative conclusion: there is no strong evidence for recursion among relative clauses, complement clauses, possessive structures and conjunctions/disjunctions. However, there is possible evidence of recursive structure in topics/repeated arguments. He also posits cultural pressures for longer or shorter sentences, such as writing systems (as I mentioned way back in 2009).

I’m sure this debate will be brought to the fore at this year’s EvoLang, with Chomsky Berwick Piattelli-Palmarini and many of the Biolinguistic crowd in attendance, and it’s a shame I’ll almost certainly miss it (unless someone wants to pay for my ticket… Just hit the donate button in the left-hand corner ).

Update: Dan Everett highlighted there are other, perhaps more exciting mediums to explore his ideas (see comments, but I still feel you can’t beat a good slideshow). So, to add some jazz to this post, below is a video of the forthcoming film he mentioned, The Grammar of Happiness:

You can read about some of the theory that I distort with my artistic license here:

Articles by Michael: Imitation in Chimpanzees ,  Animals learning syntax , Self-Domestication

Asymmetry, developmental stress and musical protolanguage (about Keelin Murray’s work)

Article by Richard: Breathing control and language

Alarm calls:  Seyfarth RM, Cheney DL, & Marler P (1980). Monkey responses to three different alarm calls: evidence of predator classification and semantic communication. Science (New York, N.Y.), 210 (4471), 801-3 PMID: 7433999

Fooling chimpanzees: Seyfarth, R., & Cheney, D. (2012). Animal Cognition: Chimpanzee Alarm Calls Depend On What Others Know Current Biology, 22 (2) DOI: 10.1016/j.cub.2011.11.050

FoxP2 and birdsong: Haesler S, Rochefort C, Georgi B, Licznerski P, Osten P, & Scharff C (2007). Incomplete and inaccurate vocal imitation after knockdown of FoxP2 in songbird basal ganglia nucleus Area X. PLoS biology, 5 (12) PMID: 18052609

Evolution of voluntary control of breathing: MacLarnon AM, & Hewitt GP (1999). The evolution of human speech: the role of enhanced breathing control. American journal of physical anthropology, 109 (3), 341-63 PMID: 10407464

Tool use: Ottoni, E., & Izar, P. (2008). Capuchin monkey tool use: Overview and implications Evolutionary Anthropology: Issues, News, and Reviews, 17 (4), 171-178 DOI: 10.1002/evan.20185

Here’s the abstract:

Social networks show striking structural regularities, and both theory and evidence suggest that networks may have facilitated the development of large-scale cooperation in humans. Here, we characterize the social networks of the Hadza, a population of hunter-gatherers in Tanzania. We show that Hadza networks have important properties also seen in modernized social networks, including a skewed degree distribution, degree assortativity, transitivity, reciprocity, geographic decay and homophily. We demonstrate that Hadza camps exhibit high between-group and low within-group variation in public goods game donations. Network ties are also more likely between people who give the same amount, and the similarity in cooperative behaviour extends up to two degrees of separation. Social distance appears to be as important as genetic relatedness and physical proximity in explaining assortativity in cooperation. Our results suggest that certain elements of social network structure may have been present at an early point in human history. Also, early humans may have formed ties with both kin and non-kin, based in part on their tendency to cooperate. Social networks may thus have contributed to the emergence of cooperation.

… Well, it sort of did and didn’t. Atkinson raised two major theoretical points in his paper. The first, and the one I’m interested in, made the link between phoneme inventory sizes, mechanisms of cultural transmission and the underlying demographic processes supporting these changes. Sadly, it was Atkinson’s second idea – that we could develop a serial founder effect model from Africa based on the phoneme inventory size – where most of the attention fell. In a methodological sense, I admired Atkinson’s approach to testing this second hypothesis, but I did feel he jumped the gun somewhat: I think more work was needed on the cultural transmission model before testing for serial founder effects. Indeed, that we haven’t developed an initial model linking the relationship between phoneme inventory size and demography, may yet prove to be Atkinson’s downfall: we should be testing multiple explanatory models (Bayesian MCMC comparison, perhaps?) rather than taking a one-size-fits-all approach.

Anyway, my long-winded and whiny intro of lost opportunities aside, the reason for this post is that Linguistic Typology has recently published a series of papers on what were probably the two big headline hitters of last year (at least in evolutionary linguistic circles): the aforementioned Phonemic Diversity paper (see here and here) and Dunn et al’s Word Order Universals paper (see here, here and here). In fact, the latter paper is a perfect example of the point I was trying to make above: they tested multiple, competing hypotheses to explain the same set of data. I’m hoping someone else will pick up on the Word Order commentary (and that includes anyone outside the usual contributors, so here’s my email), because I’m not sure whether I’d do the topic justice. Over the coming weeks, however, I will dedicate several posts to the great vanishing phoneme debate, with each post gradually laying out my theoretical model for linking cultural transmission and socio-demographic forces with the size of a language’s phoneme inventory.

For now, I’ll just provide a brief list of the authors (and their papers) that critique Atkinson’s original article, which, if you know your linguists, pretty much reads like an A-list of the big names in the field (click here for the link to the issue):

• Joan Bybee: How plausible is the hypothesis that population size and dispersal are related to phoneme inventory size? Introducing and commenting on a debate.
• Peter Trudgill: Social Structure and phoneme inventories.
• Mark Donohue and Johanna Nichols: Does phoneme inventory size correlate with population size?
• Ӧsten Dahl: Are small languages more or less complex than big ones?
• Søren Wichmann, Taraka Rama, Eric W. Holman: Phonological diversity, word length, and population sizes across languages: The ASJP evidence.
• Richard Sproat: Phonemic diversity and the Out-of-Africa theory.
• Claire Bowern: Out of Africa? The logic of Phoneme Inventories and Founder Effects.
• Vladimir Pericliev: On phonemic diversity and the origin of language in Africa.
• Don Ringe: A pilot study for an investigation into Atkinson’s hypothesis.
• Keren Rice: Athabaskan languages and serial founder effects.
• Bill Ross and Mark Donohue: The many origins of diversity and complexity in phonology.
• Ian Maddieson, Tanmoy Bhattacharya, Eric D. Smith and William Croft: Geographical distribution of phonological complexity.
• Florian T. Jaeger, Peter Graff, William Croft and Daniel Pontillo: Mixed effect models for genetic and areal dependencies in linguistic typology.
• Quentin D. Atkinson: Linking spatial patterns of language variation to ancient demography and population migrations.

I’ll hopefully mention some of these in the coming weeks, but I’ve only got limited access to them (apparently, even institutional access is not enough to bypass the gated walls of this archaic system of distributing knowledge).

References
Atkinson, Q. (2011). Phonemic Diversity Supports a Serial Founder Effect Model of Language Expansion from Africa Science, 332 (6027), 346-349 DOI: 10.1126/science.1199295

N.B. The featured picture is of the painting The Boyhood of Raleigh by John Everett Millais. The story behind its use here is simple: it appears on the storytelling page of wikipedia. It also began me on a quest of hyperlink jumps, but that’s another story… Yawn.

On an only tangentially related note, I’m still trying to decide if this is the use of a tool or not. If so, it’s a major win for dogs.

The mushroom, according to McKenna, had also given humans their first truly religious experiences (which, as he believed, were the basis for the foundation of all subsequent religions to date). Another factor that McKenna talked about was the mushroom’s potency to promote linguistic thinking. This would have promoted vocalisation, which in turn would have acted in cleansing the brain (based on a scientific theory that vibrations from speaking cause the precipitation of impurities from the brain to the cerebrospinal fluid), which would further mutate the brain. All these factors according to McKenna were the most important factors that promoted evolution towards the Homo sapiens species. After this transformation took place, the species would have begun moving out of Africa to populate the rest of the planet. Later on, this theory by McKenna was given the name “The ‘Stoned Ape’ Theory of Human Evolution”.

I’m fairly sure this just offloads part of the craziness onto McKenna, but I might name drop the theory next time I’m looking for a more lively reaction when discussing language origins.

N.B. This is one of my many posts that was written some time ago. I decided to publish in 2012 just in case the customer was right about our impending doom. With that out of the way, we can now focus on the critical issues surrounding the size of a language’s phoneme inventory and the distribution of Psilocybe cubensis.

“1. Mirror self-recognition

2. Tests of metacognition;

3. Metacognition of others’ mental states” (Gómez 2009: 45)

After having discussed the first capacitiy in my previous post, I will discuss the latter two in this post, starting with metacognition, that is being aware of one’s own knowledge states, and then turn to being aware of other’s mental states.

Metacognition.

Being aware of one’s own mental states, i.e., reflective consciousness, surely seems to be one of the most crucial components of self-awareness. In one paradigm used to test for metacognitive awareness, monkeys were trained to select, out of a number of two or more images, the one that is identical to an image they have been shown earlier. As is to expected, the monkeys’ performance progressively deteriorated the longer the delay was between the sample image and the selection task.

source

After they had learned how the task worked, the monkeys were then given a choice between performing the task or not. If they chose to perform the task, they had the chance of getting a preferred reward when right or nothing when wrong. If they chose not to do the task they simply got a low quality reward. Interestingly, when they chose to perform the test their success rate was significantly higher than when they opted for not doing it but were forced to perform the test nevertheless.

These results have been taken as evidence that monkeys are aware of their own knowledge states and base their decision on whether to take the test or not on some form of reflexive consciousness or metacognition. One significant problem of this setup, however, is that in principle there could be a wealth of different possible explanations for these behaviours. For example, the monkey’s decision could be a result of a low-level uncertainty monitoring mechanism that regulates its behaviours without the animal necessary having a conscious, second-order awareness of its own mental states (Gómez 2009: 46f.).

Similar experiments have also been conducted with dolphins, who were presented with different sounds and had to push one of three levers (one for ‘high’ tones, one for ‘low’ tones, and one if they were ‘uncertain’) in response. The dolphins tested performed similar to humans tested on the same task, and given what we know about dolphin cognition and encephalization in general it seems feasible to interpret these results as evincing self-awareness (Herman 2009: 44).

Metacognition of Others’ Mental States/ Mentalizing

The question whether chimpanzees have a ‘theory of mind’, that is, whether they are conscious of the mental states of others, has been a controversial topic for more than thirty years (Blackmore 2010: 214f., Hare 2011: 294ff.). Until quite recently, there was a broad consensus that theory-of-mind abilities and the ability for mentalizing, defined as the

“implicit or explicit attribution of mental states to others and self (desires, beliefs) in order to explain and predict what they will do” Frith & Frith 2012)

were uniquely human. There was an overall consensus that there was a

“vast qualitative gap […] between the social cognition of human and non-human apes” (Hare 2011: 296).

This view is echoed for example in the statements by Penn et al. (2008) mentioned in my first post and also by Susan Blackmore (2010: 215), who reports on experiments which showed that when begging for food, chimpanzees did not discriminate between an experimenter who could see them and one who could not (e.g. because s/he had a bucket over her/his head) but randomly gestured to both of them. The conclusion from this experiment was that chimpanzees had no concept of seeing, something which human children develop as early as in their second year of life.

However, this view has changed dramatically over the last ten years, when researchers started to test chimpanzees in competitive, as opposed to cooperative contexts, which are much more natural and ecologically valid for them. In competitive paradigms, chimpanzees do indeed show an understanding of seeing, e.g. by taking the piece of food that is hidden from the view of a dominant competitor instead of the piece of food seen by both (Hare 2011: 298f.).

source

What is more, chimpanzees even know that others make inferences. For example, in the context of searching for a single piece of food, chimpanzees were presented with two opaque boards on a table, one lying flat, and one slanted. Chimpanzees normally chose the one that is slanted because there could be food lying under it. However, when there was another chimpanzee there before them, they tend to avoid the slanted board, because they assume that their competitor has already looked under it guided by the same inference they themselves would make (Schmelz et al. 2011).

This, taken together with a wealth of other recent results on the sophisticated sociocognitive abilities of chimpanzees, suggests that chimpanzees are conscious of at least some psychological states in others, although not to the same degree as humans, who from early age on outperform all other animals in every regard (Hare 2011: 300f.; Carpenter 2011, see also this post). It should be noted here that these conclusions not only extend to great apes, but that similar capacities have also been demonstrated, for example, in  ravens (Blackmore 2010: 215). As Hannah wrote last month (see here), they even point declaratively to conspecifics with their beaks in order to show them something (Pika & Bugnyar 2011), a behaviour made even more remarkable by the fact that it is not found in great apes at all (Tomasello 2006, although this conclusion is hotly debated see, e.g. here).

source

As my two posts should have illustrated, it is far from clear that Darwin really made a mistake when speaking of the difference between human and nonhumans as “one of degree, not of kind.” This does not only hold for cognition. Instead, the experiments I’ve written about here, taken from the domains of mirror self-recognition, metacognition of one’s own mental states and metacognition of the mental states of others, suggest “that animals other than humans are conscious and have subjective experiences that rely on some degree of consciousness” (Burkhard & Bekoff 2009: 42).

References:

Blackmore, Susan (2010): Consciousness: An Introduction. 2nd Edition. Oxford: OxfordUniversity Press.

Burkhardt, Gordon M. and Marc Bekoff (2009): “Animal Consciousness.” In: Tim Bayne, Axel Cleeremans and Patrick Wilken (eds.): The Oxford Companion to Consciousness.Oxford:OxfordUniversity Press, 39-43.

Carpenter, Malinda (2011): “Social Cognition and Social Motivation in Infancy.” In: Usha Goswami (ed.): The Wiley-Blackwell Handbook of Childhood Cognitive Development. 2_nd Edition. Malden, MA [et al.]: Blackwell, 106-128.

Frith, Chris D. and Uta Frith (2012): “Mechanisms of Social Cognition.” In: Annual Review of Psychology 63: 287-313.

Gómez, Juan-Carlos (2009): “Animal Consciousness: Great Apes.” In: Tim Bayne, Axel Cleeremans and Patrick Wilken (eds.): The Oxford Companion to Consciousness.Oxford:OxfordUniversity Press, 45-48.

Hare, Brian (2011) “From Hominoid to Hominid Mind: What Changed and Why?” In: Annual Review of Anthropology 40(1): 293-309.

Herman, Louis M. (2009): “Animal Consciousness: Dolphins.” In: Tim Bayne, Axel Cleeremans and Patrick Wilken (eds.): The Oxford Companion to Consciousness.Oxford:OxfordUniversity Press, 43-45.

Penn, Derek C, Keith J. Holyoak and Daniel J. Povinelli (2008): “Darwin’s Mistake: Explaining the Discontinuity between Human and Nonhuman Minds.” In: Behavioral and Brain Sciences 31.2: 109-130.

Pika, Simone and Thomas Bugnyar (2011): “The Use of Referential Gestures in Ravens (Corvus Corax) in the Wild.” In: Nature Communications 2: 560.

Schmelz, Martin, Josep Call and Michael Tomasello “Chimpanzees know that Others Make Inferences.” In: Proceedings of the National Academy of Sciences 108: 17284-17289.

Tomasello, Michael (2006): “Why Don’t Apes Point?” In: Nick. J. Enfield and Stephen C. Levinson (eds.): Roots of Human Sociality: Culture, Cognition and Interaction. Oxford: Berg, 506-524.

For those of you who know students or colleagues who might be looking for an introduction to NLP next quarter, encourage them to join us and the 40,000 students who have already registered in the course!

Students have access to screencast lecture videos, are given quiz questions, review exams and programming assignments in Java or Python, receive regular feedback on progress, and can participate in a discussion forum.

The course covers a broad range of topics in natural language processing at the advanced undergraduate or introductory graduate level, including word and sentence tokenization, text classification and sentiment analysis, spelling correction, information extraction, parsing, meaning extraction, and question answering, We will also introduce the underlying theory from probability, statistics, and machine learning that are crucial for the field, and cover fundamental algorithms like n-gram language modeling, naive bayes and maxent classifiers, sequence models, probabilistic dependency and constituent parsing, and vector-space models of meaning.

You can find more information about joining at http://www.nlp-class.org/