The Adventures of Auck

A fun Language Experiment!

2010-12-08T11:41:00.003Z

I'm running a pilot experiment - you can help me by taking part - it only takes 5 minutes!

It's a memory experiment where you'll hear an alien language, then answer questions on it.

Take part here

Learning the Form of Causal Relationships Using Hierarchical Bayesian Models

2010-11-19T17:42:00.006Z

The mutual exclusivity bias has been posited as a fundamental learning bias (Markman & Wachtel, 1988; Merriman & Bowman, 1989). However, there is mounting evidence that bilinguals do not exhibit mutual exclusivity (Merriman & Kutlesic, 1993; Byers-Heinlein & Werker, 2009;Healey & Skarabela, 2009; Houston-Price, Caloghiris & Raviglione, 2010). It is hypothesised that the amount of variation in the input of bilinguals either lacks enough evidence for a mutual exclusivity bias to emerge, or renders it ineffective. Previously, I showed that a Bayesian model of cross-situational learning (Frank et al., 2009) could not account for differences found between the mutual exclusivity behaviour of monolingual and bilingual children. A reasonable step towards capturing this behaviour would be to add a higher level of abstraction to the model. This would allow the model to alter the strength of its own mutual exclusivity bias in accordance with the amount of variance it encountered.

The research described here implements a hierarchical Bayesian model of causal structure (Lucas & Griffiths,2010;Lucas, 2010). Although it does not address mutual exclusivity directly, it is very relevant. Adult's and children's inferences about causal relationships were affected by their previous experience. A hierarchical Bayesian model that could adjust its assumptions about causal structure was shown to match the performance of adults.
When we learn about causal relationships, we learn to associate a cause with an effect. When there are multiple possible causes, we also need to consider the functional form of the relationships. For example, imagine that two switches are connected to a light. There are a number of possibilities that could cause the light to turn on. Perhaps only one switch needs to be activated (OR), perhaps both are needed (AND). The switches may always work the same (deterministic) or vary some proportion of the time (stochastic or `noisy').

Lucas & Griffiths (2010) show that a hierarchical Bayesian model can match the behaviour of adults in a learning task where they had to make inferences about causal relationships. Previous models of causal structure learning either assume particular functional forms of causal relationships, making them inflexible, or make no assumptions, rendering them incapable of capturing effects of context (see Griffiths & Tenenbaum, 2005, Lucas & Griffiths, 2010). The hierarchical model infers the functional form of the causal relationship as well as the exact relationship between variables.

Although not couched in terms of cross-situational learning, the task is compatible with one. Participants were shown wooden blocks, identical except for a one-letter label (A, B and C in the figure above). The task was to learn which were `blickets'. To help them, there was a `blicket meter' - a device that activated when in the presence of a blicket. Participants were shown several training rounds where one or two blocks were placed on the blicket meter and observed the meter activating or not. After training, they saw another set of blocks (D, E and F) go through a series of blicket tests. They were asked to indicate how confident they were that each block was a blicket.

The experiments had different training conditions. In one, the training data was consistent with the blicket meter's response having a disjunctive relationship (OR) with its causes. That is, it activated when any of the blocks placed on meter were blickets. In another condition, the blicket meter responded consistently with a conjunctive relationship (AND). That is, two blickets were required to activate the blicket meter. The test block was set up to be consistent with either training condition.

Participants' responses in the test block were affected by their experience during training. They saw block D fail to activate the meter 3 alone 3 times, block E fail once and blocks D and F together activate the meter twice. If you assume a disjunctive relationship, then D failing 3 times should be evidence against D being a blicket, while E failing once is less evidence. Indeed, participants in this condition rated D as being less likely to be a blicket than E. Assuming a conjunctive relationship, however, D failing to activate the meter is not informative, whereas seeing D and F activate the meter together is evidence for D being a blicket. Participants given conjunctive training rated D as being more likley to be a blicket. The model matched the participants responses closely.

Lucas & Griffiths argue that this shows that people can make inferences appropriate to causal relationships with more than one kind of functional form (e.g. conjunctive, disjunctive) and that their inferences can be affected by evidence `transferred' from a previous experience. In other words, participants' assumptions about causal relationships can be modified by experience, and this can lead to qualitatively different behaviour.

Lucas (2010) also shows that the model accounts for children's behaviour, too. However, children's responses were more affected by the likelihood than the adults, while adults tended to assume an OR function. This suggests that children are more flexible learners.

However, I'm unsure whether hierarchical Bayesian modelling can be applied to language. Causal forms and causal relationships have a definite hierarchy. But what about F1 and F0? English uses formants to make distinctions at the lexical level and pitch to make distinctions at the pragmatic or phrasal level. Tonal languages, however, use pitch at the lexical level, and some have morphological markers of phrasal boundaries (see Black, 2000).

I suggest that Bayesian models will always have built-in assumptions about the structure of the phenomenon. In studying language evolution, we should be focussing on how this structure emerges in the first place. I propose a different kind of hierarchical model that does not specify a structure in advance. Rather, the role of each level of the hierarchy should be determined by the data. This should be based on the most salient cues that divide the variance in the data in the most functional way.

One possible solution could be general hierarchical dynamic expectation maximisation models. But more on this in the future.

Lucas, C. G., & Griffiths, T. (2010). Learning the Form of Causal Relationships Using Hierarchical Bayesian Models Cognitive Science, 34 (1) pdf

Lucas, C. G. (2010). Developmental differences in learning the form of causal relationships Proceedings of the Cognitive Science Society pdf

How to fix Java for Firefox 3.6 on Mac

2010-08-25T15:07:00.004+01:00

A few days ago I updated Firefox and found that java wasn't working. This is a problem, since I'm developing java programs.

The Software Update facility could not find an update for java or Firefox. Although there were some solutions on the net, none of them worked.

In the end, I reinstalled the Java Embedding Plugin in the following way:

Download and unzip the Java Embedding Plugin.

Follow the Read-Me, reproduced here:

1) In the JavaEmbeddingPlugin folder you just downloaded and unzipped, open the Binaries folder and drag JavaEmbeddingPlugin.bundle and MRJPlugin.plugin to "/Library/Internet Plug-Ins" folder.

2) If you're running Mac OS X 10.4.X (Tiger), you also need to make
sure MRJPlugin.plugin's timestamp is later than the timestamps
of two other files in the "/Library/Internet Plug-Ins" folder --
"Java Applet.plugin" and "Java Applet Plugin Enabler". This
isn't necessary on Mac OS X 10.5.X (Leopard).

Open a Terminal session and enter the following command:

touch "/Library/Internet Plug-Ins/MRJPlugin.plugin"

3) Find your browser in the Applications folder (e.g. Firefox), Control-click (or right-click) on your browser's binary and choose "Show Package Contents".

4) Browse to the Contents/MacOS/plugins folder and delete JavaEmbeddingPlugin.bundle and MRJPlugin.plugin.

Language Evolution and Tetris!

2010-08-06T10:41:00.000+01:00

Hello, people of the Blogosphere!

Why not take some time out from your dedicated reading to do a little language evolution experiment! And all you have to do is play Tetris!

... and learn an alien language. It takes no less than 10 minutes.

The instructions and game are here:

http://blake.ppls.ed.ac.uk/~s0451342/tetris/Tetris_Experiment.htm

Due to me being a terrible programmer, it'll probably crash or do some weird things. But it's all in the name of pseudo-science!

I'm blogging at Replicated Typo

2010-07-26T15:50:00.001+01:00

I am now blogging at A Replicated Typo - don't expect to see to much more on this blog!

Cultural Induciton is Hard

2010-06-24T19:08:00.004+01:00

Chater and Christiansen (2010) argue that culturally transmitted systems such as language are easier to learn than natural systems because they have adapted to learner's biases, so their intuitions will likely be correct. Being a speaker of one of the morphologically most complex languages in the world, I'm not so sure...

Maggie Tallerman gave a keynote speech at this year's EvoLang conference. A section of it used grammatically acceptable and unacceptable sentences in Welsh to illustrate the point. As a somewhat lapsed Welshspeaker, whose knowledge of Mutation was never great, I was a bit embarrassed to find I wasn't sure if the examples were correct. Last month, too, Mike Dowman gave a talk at Edinburgh University emphasising how impressive it is that we all make the same grammaticality judgements about sentences we have never seen before.

I've long wondered whether this is the case with Welsh mutation. Consonant mutation, or 'Treiglo' in Welsh, occurs in many Celtic languages and is a terrible affliction for the second language learner. In a number of (grammatical) contexts, the initial consonant of a word (nouns and verbs) changes to another. For instance, 'kitchen' in Welsh is 'cegin' [k3gIn], but 'his kitchen' is 'ei gegin' [g3gIn] and 'her kitchen' is 'ei chegin' [X3gIn].

There are three forms of mutuation - soft, nasal and aspirate. The Wikipedia page on Welsh Mutation gives a broad overview. The contexts they apply in are extensive, for example:

Nouns after the preposition 'in'
After imperatives.
After the personal pronoun (my)
Singular feminine nouns after the definite article (but not words beginning in ll or rh)
In the negative form of verbs in the Short Future Tense
Masculine nouns after 'three' and all nouns after 'six'

The rules of mutation in old Welsh were much simpler: It only occured for feminine nouns after the definate article. However, presumaply by a process of analogy, the 'rules' spread and became more complex. And here's the point I'm trying to make: Welsh morphology may be so complex, and subject to so much change by analogy, that there is very little agreement between people.

I'm not saying that mutuation is unsystematic. There's even an automatic mutuation checker online. However, it always annoys me when syntacticians cite some examples that they haven't actually gone out and tested.

Now, this may be an intuition I have from school. I grew up speaking Welsh, both my parents speak Welsh and I went to a Welsh-medium nursery, primary and secondary school where we were not allowed to speak English. Despite this, I'm not a confident speaker, especially after 7 years outside of Wales. I was particularly bad at mutation (although my English spelling was, and still is, equally as bad). We had it drilled into us with tables and excersises, but I still can't really do it properly. This is partly because of the minority langauge status of Welsh, and the fact that everybody spoke English as a form of rebellion. The influence of English has been felt in other areas of Welsh such as Subject-Verb order, too.

However, I've always felt guilty about not being able to speak the mothertongue properly. Then I became a linguist and found a way out: All along, my teachers had been prescribing language, and that prescription was a few generations old. In terms of language evolution, the language the children speak is the correct language (the descriptivist approach).

Long story short, I conducted my own grammaticality judgement experiment. I couldn't find a grammaticality judgement for Welsh mutation online, nor any information about how well learners pick it up (please send me links if you know of any!). Neither am I a trained syntactician, so I have no real idea how to do an experiment, nor do I have access to money to employ participants.

So I decided to do it in the form of a facebook quiz, using Quibblo. I found example sentences from an instructional pamphlet and took one from each major context. I then created alternative mutations for each sentence. Participants were presented with an English equivalent of each sentence, and asked to indicate which sentences they thought were correct. Participants could make more than one choice.

You can take the quiz here and view the results here.

12 people participanted, mainly schoolfriends since this was distributed via facebook. This is a good point rather than a bad point, since we're more likely to have been exposed to the same linguistic environment (and indeed been in frequent contact). However, many of these may be, like me, somewhat out of practice. On the other hand, 11 indicated they were 'fluent' and 1 'intermediate' speakers of welsh. All but one came from South Wales. All particpants chose only one answer in 17 out of 34 questions.

As it turned out, Quibblo wasn't a very good choice - it only records totals, not individual records of participant's choices. Anyway, here's some analysis.

For each setence, I worked out the average agreement. This is the likelyhood of any two people agreeing that at least one form was correct. For all sentences, the average agreement was 67.1% For sentences where participants only selected one answer, the average agreement was 60.9%.

Some sentences recieved 100% agreement - for the sentence 'the boy', all 12 participants chose the prescribed 'dy fachgen'. However, the sentence 'the girl' was split with two thirds going for the prescribed 'y ferch' and one third going for 'y merch'. On the other end of things, 9 participants chose 'dydd mawrth' for the meaning 'on tuesday', when the prescribed form is 'ddydd mawrth', which only 3 participants chose.

For the setences with more than two possible choices, the choices are spread. For the sentence 'I read a good book', 6 different options were selected with an average agreement of 54.5%. The worst agreement was for the sentence 'the sixth girl' with participants agreeing on average 25.5% of the time between 4 options. For this sentence, 12 participants chose 16 options, meaning that some participants thought at least two options were correct (I don't have the exact data on who chose what).

I put some tricky questions in to see what would happen. The first was designed to test whether adjacent adjectives should be mutated. That is, adjectives which follow a singular feminine noun mutate, but it's not clear whether a following adjective should too. Participants were given the sentence 'a big, tall, good girl' and given the option to mutate none, the first adjective, first and second adjectives and all adjectives. 3 participants chose to mutate only the first while 5 chose to mutate all adjectives (3 chose to mutate none and 1 chose to mutate two). The agreement was 24.2%.

The second tricky question involved loanwords. Nouns after a conjunction mutate, so participants were given the sentence 'gin and tonic' and the options 'jin a tonic' and 'jin a thonic'. Agreement was slightly better here at 84.8% in favour of the prescribed (and attested) mutated form, although one person thought both were correct.

The final one involved analogy and loanwords. I heard someone mutate 'chips', which has a [ tʃ ], which doesn't exist in Welsh to the voiced equivalent [ dʒ ]. There is no prescription here, but this makes perfect sense if mutation really does spread by analogy. Presented with the sentence 'a bag of chips', 5 participants voted for the unmutated variant and 10 for 'bag o jips' (83.3% agreement).

All in all, enough to make my old Welsh teachers weep. Of course, the sample is probably skewed and can't be verified and some might have looked the answer up etc. But part of the point of this is that, for very simple sentences, people should be choosing the same sentences.

In a forthcoming special issue of Cognitive Science (preview here), Nick Chater and Morton Christiansen argue that learning cultrually-transmitted systems is easier than learning about the natural world because cultural systems will be adapted towards a learner's biases. Therefore, a learner's intuitions and guesses are likely to be correct. That is, it's easier to co-ordinate your behaviour with other people than it is to be right about the world (an alternative name for the paper could have been 'Language Evolution: Specifically not the hardest problem in Science').

It's a great paper, and argues for my PhD thesis - that language acquisition should be looked at in the light of language evolution. However, cultrual induction may not be easier than learning about the natural world if everybody is doing something different. Consider the participants in my experiment: A child learning from them faces sources of cultural variants that not only contradict each other half of the time, but contradict themselves part of the time. At least mass and other physical attributes are Universal - gravity doesn't work differently in North Wales. However, since the data cultural learners are presented with comes from multiple people who themselves may have had different and non-overlapping sources of input, cultural learning may be pretty tricky after all.

So, there may be space in this topic for my PhD thesis: The social structure of cultural learners will have a huge impact on the ease of Cultural induction, and thus on the pressures and eventual forms of language.

Nick Chater & Morton H. Christiansen (2010). Language Acquisition Meets Language Evolution Cognitive Science

Evolutionary approaches to Bilingualism

2010-05-25T13:11:00.004+01:00

I recently gave a talk at the University of Edinburgh LEL Postgraduate Conference. It was my first ever talk and it really forced me to figure out what I'm supposed to be studying! Here's a video of my talk:

Frank MC, Goodman ND, & Tenenbaum JB (2009). Using speakers' referential intentions to model early cross-situational word learning. Psychological science : a journal of the American Psychological Society / APS, 20 (5), 578-85 PMID: 19389131

Hunag, Y. (2009). Supporting Meaningful Social Networks Technical Report, ECS, University of Southampton
pdf

Healey, E. and Scarabela, B. (2009). Are children willing to accept two labels for one object? Proceedings of the Child Language Seminar. University of Reading.

Byers-Heinlein K, & Werker JF (2009). Monolingual, bilingual, trilingual: infants' language experience influences the development of a word-learning heuristic. Developmental science, 12 (5), 815-23 PMID: 19702772

E-Coli, Linux and Language

2010-05-13T22:17:00.006+01:00

A recent post on The Loom looks at a paper by Koon-Kiu Yang et al. which compares the hierarchical structures of the operating system Linux and the bacterium E-Coli. Really interesting analysis - and a good discussion on the blog.

I found it interesting that E-coli's structure is primarily lower-level 'workhorses' with relatively few master controllers. Linux on the other hand has a much larger percentage of high-level 'master' and 'middle manager' modules and reletively few 'workhorses'. Linux is designed while E-coli is evolved.

I’m wondering how linguistic systems would fit into this schema. What are the ‘workhorses’ and ‘master regulators’ of language? There are many more ‘low-level’ words that refer to things than ‘higher level’ syntactic structures. This would make it like e-coli.

On the other hand, there are relatively few ‘low level’ phonemes and very many ‘high level’ concepts. This would make it more like Linux.

Maybe language has more ‘middle managers’ than anything else?

Answering this may give an insight into how ‘designed’ language is, as opposed to ‘evolved’.

Yan, K., Fang, G., Bhardwaj, N., Alexander, R., & Gerstein, M. (2010). Comparing genomes to computer operating systems in terms of the topology and evolution of their regulatory control networks Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.0914771107

Mutual Exclusivity biases in cross-situational learning: A comparison between monolingual and bilingual corpora

2010-05-11T14:30:00.000+01:00

This report focuses on models of cross-situational learning and how current models compare when exposed to real monolingual and bilingual input. Several model types were evaluated against two transcribed videos of parent-child interaction, one being monolingual and the other being bilingual. Children have been shown to demonstrate a Mutual Exclusivity (ME) bias (Markman and Wachtel, 1988) during word learning. Frank et al. (2009) showed that their model also exhibited Mutual Exclusivity (ME) behaviour after learning from a monolingual corpus of contexts. The current study takes the same model but with bilingual input and asks whether the same behaviour is exhibited.

Method

Frank et al. (2009) provide a transcribed video of monolingual parent-child interaction coded for use in cross-situational learning. An equivalent bilingual corpus was looked for. The main criterion was a roughly equal number of utterances in both languages. The CHILDES database has suitable resources. A recording from a study by Yip and Matthews was selected (see CHILDES, 2010). The child in question was a native bilingual from birth. Her mother was a native Hong-Kong Cantonese speaker and her father was a native speaker of British English. She was 2;11 in the chosen recording. There are 967 utterances, 48% of which are Cantonese and 52% are English. The objects visible in the video were added to the transcription, along with the mapping between words and objects and the referential intentions of the speakers. The coding scheme was adopted from Frank et al. (2009). The code for the models was supplied by Frank et al. (from Frank,2010).

Results

The performance of different models were analysed in two ways. Firstly, the best estimated lexicon (word-object mappings) of each lexicon was evaluated against a gold-standard lexicon in terms of precision, recall and the resulting F-score.Secondly, the models were asked to guess the intended referent of each utterance-object context.Tables 1 and 2 show the lexicon results for the monolingual and bilingual corpora respectively. Frank et al.’s model returns the highest F-scores in both cases.This is largely due to an advantage in precision, likely stemming from the modeling of non-referential words. Frank et al.’s model returned a word-object mapping for the bilingual corpus with a precision of 0.31 and a recall of 0.27, giving an F-score of 0.29. This is lower than the score for the same model on the monolingual corpus. This could be due to the referential uncertainty (independent from amount of synonymy) in the bilingual corpus being higher.The results for the referential intentions, shown in Tables 3 and 4, have different trends. For the monolingual case, the precision of Frank et al.’s model allows it to outperform the other models. However, it performs relatively poorly in the bilingual case, with the Conditional Probability model performing best. However,all models perform with little precision and recall, suggesting that the task is harder. With more data, results might be different.

Mutual Exclusivity

After the model had processed the corpus, it was presented with a mutual exclusivity task and the relative likelihood of several interpretations were measured.In the task, the model was presented with a context with a new object (e.g. a dax) and a familiar object (a bird for the monolingual model and an orange for the bilingual model) and a new word (e.g. ”dax”). The probabilities were calculated for the model linking the new word with neither object (i.e. it considers the word non-referential), linking it with the new word, linking it with the old word or linking it with both. Figure 1 shows the results of the task with the results for a ’monolingual’ model for comparison.

The monolingual results are re-calculated for this study, so differ slightly from those reported in Frank et al. (2009).The results for the monolingual and bilingual models have the same trend - both rank the possible situations in the same order of likelihood. The most likely situation is the new word being linked to the new object, honouring mutual exclusivity. The second most likely situation is that the word refers to neither object.Intuitively, one would expect a bilingual to be more likely to consider that the new word was another word for the familiar object. Indeed, the bilingual model does consider this possibility relatively more likely than the monolingual model. However, the model still considers neither mapping to be more likely than an extra synonym. This may mean that, given an additional cue (e.g. pragmatic), the bilingual would be more ready to accept a synonymous interpretation. This is an empirical question.

The Prior Probability

The prior probability is simply the number of mappings in the hypothesised lexicon, modulated by a fixed parameter (alpha). This represents a preference for smaller lexicons. This means that a hypothesis which results in the lexicon with fewest mappings will receive the highest prior probability. With the default parameter (alpha = 7), the Mutually Exclusive preference (for DAX-dax) beats the preferences for the original mapping (map neither word to the unfamiliar object),both mappings and the mapping of the unfamiliar object with the familiar name.However, this ranking depends on the lexicon size bias (alpha) parameter. With a low alpha, the most likely mapping is the ME mapping. With a higher alpha, the most likely mapping is the original mapping (see Figure 2).

The same trend also exists between the preference for the ME mapping and both mappings, although the preference for both mappings does not overtake the preference for the ME mapping (see Figure 3).
The explanation is as follows: The original mapping receives a high prior probability because it doesn’t increase the size of the lexicon. However, the likelihood of experiencing a non-referential word is low, leading to a total probability that favours the ME mapping over the original. Assuming a larger lexicon (decreasing alpha), the relative increase in lexicon size is smaller, tipping the balance between the original and ME mapping preferences.Interestingly, the likelihood of choosing both mappings overtakes the original mapping when alpha is less than 1 (see figure 4).
That is, the likelihood of assuming both mappings increases when the prior is set to less than the number of word-object mappings in the lexicon. Such a setting makes sense for a bilingual (who have up to twice as many mappings as bilinguals) because it represents the number of concepts. Put another way, by compensating for the additional synonymy in bilingual input, the likelihood of assuming both mappings increases.The dependence of the ME experiment results on alpha is acknowledged by Frank et al.:

“Note that there is some parameter dependence in our models fit to the mutual exclusivity situation. Depending on the size of the corpus,it might be the case that the prior disadvantage of adding a word to the lexicon would not be outweighed by the increase in corpus likelihood caused by learning a new word. This fact makes a developmental prediction: in early development, when very few words are known,inferences about mutual exclusivity should be weaker.”
Supporting Information for Frank et al. (2009), p. 13.

This prediction is borne out in some studies (Merriman and Bowman, 1989; Frankand Poulin-Dubois, 2002; Merriman et al., 1993). However, Markman and Wachtel (1988) found that the ME constraint weakens over time, with older children showing less of a bias, while Deak et al. (2001) find no change.The issue here is the size of the lexicon. Bilingual children may know more words than monolinguals, but it may be more accurate to judge the lexicon size by the size of one language’s lexicon.The model does not provide a mechanism for modulating the lexicon size prior parameter during learning. Currently the prior is modulated by the alpha parameter and the number of mappings, meaning that adding new mappings is dis-preferred. Bilinguals will have a higher number of mappings, altering their prior probabilities. However, this does not lead to qualitative differences in the mutual exclusivity experiment.The motivation for modulating the prior by the number of mappings is mainly to simplify the model.

“We chose a prior probability distribution that favored parsimony,making lexicons exponentially less probable as they included more word-object pairings ... The choice of a simple prior puts most of the work of the model in the likelihood term ... hence, the likelihood term captures the learners assumptions about the structure of the learning task.”
Frank et al., 2009, p. 579

That is, the decision is driven by the statistical, computational approach to the formal problem rather than being psychologically motivated. Therefore, the interpretation is that mutual exclusivity behaviour stems from the child’s unwillingness to learn new signal-meaning mappings. This seems a little circular - children prefer not to extend mappings from familiar words to unfamiliar objects because they prefer not to extend mappings. It also seems to go against children’s obvious ability and motivation for learning new words and meanings. Several solutions which would make the prior more sensitive to the input involve incorporating the number of concepts, the number of words or the amount of synonymy (proportional to the number of words in the lexicon divided by the number of concepts). However, the nature of the model now changes - we are using it to test specific hypotheses about mutual exclusivity, judged against empirical data,rather than seeing if mutual exclusivity ’falls out’ of more basic assumptions.

Concept-based Prior

The mapping-based prior was biased towards a monolingual mode. The model was altered so that the prior was negatively related to the number of objects in the lexicon. This represents the number of concepts for which the child knows words. The model was run on the bilingual corpus and returned a lexicon with a precision of 0.05, a recall of 0.41 and a resulting F-score of 0.09. The model was also run on the monolingual corpus again, returning a precision of 0.05, a recall of0.79 and an F-score of 0.09. For both monolingual and bilingual corpora, the recall of this model is better than for a mapping-based prior, but the precision is much worse. That is, the model overestimates the number of word-concept mappings. In fact, the models accumulated many hundreds of word-concept mappings for tens of objects (Monolingual: 551 mappings for 22 objects and 419 words; Bilingual:641 mappings for 55 objects and 598 words). The models have failed to acquire a useful vocabulary.However, running the Mutual Exclusivity experiment again, the relative ranking of the preferences has changed. Although the ME mapping is still favoured, the next preferred interpretation is to make both mappings (rather than neither, see Figure 5). However, this difference is exhibited with both monolingual and bilingual input data. By neutralising the difference in the prior, the corpus likelihood now plays a bigger role, leading to a difference in the preferences.

How ’Monolingual’ is the Monolingual corpus?

Although the monolingual corpus is taken from a carer speaking one language, the lexicon the model learns contains synonymy. In fact, for the 15 objects it learned words for, 8 had more than one associated word. For half of these 8 objects, all synonyms were appropriate (e.g. ’bird’ and ’birdie’ to describe the object ’duck’),but half were not appropriate. In other words, the model accommodates synonymy.The original Mutual Exclusivity experiment in Frank et al. was done with the object ’bird’, which had one associated word. The ME experiment was applied for all words that the model learned from the monolingual corpus. There were no significant differences between the posterior probabilities for any of the situations (DAX-dax, Both etc.) for synonymous mappings versus non-synonymous mappings. This holds for both the original and the concept-based prior.

Conclusion

Frank et al.’s model can be used to model word learning in bilinguals. There are some quantitative differences in the ME behaviour of models run on monolingual and bilingual corpora. However, no qualitative differences were found. Even when the prior bias for minimising the number of mappings was neutralised, both models still preferred to map the new object with the new word.

Next Steps
The results are inconclusive, but may reﬂect the limited data. I suggest that synthetic corpora would make the dynamics more clear. Very simple cross-situational learning corpora could be created with varying amount of ’bilingualism’.

References

Frank MC, Goodman ND, & Tenenbaum JB (2009). Using speakers' referential intentions to model early cross-situational word learning. Psychological science : a journal of the American Psychological Society / APS, 20 (5), 578-85 PMID: 19389131

Byers-Heinlein K, & Werker JF (2009). Monolingual, bilingual, trilingual: infants' language experience influences the development of a word-learning heuristic. Developmental science, 12 (5), 815-23 PMID: 19702772

Deák GO, Yen L, & Pettit J (2001). By any other name: when will preschoolers produce several labels for a referent? Journal of child language, 28 (3), 787-804 PMID: 11797548

Frank, I., & Poulin-Dubois, D. (2002). Young monolingual and bilingual children's responses to violation of the Mutual Exclusivity Principle International Journal of Bilingualism, 6 (2), 125-146 DOI: 10.1177/13670069020060020201

Markman EM, & Wachtel GF (1988). Children's use of mutual exclusivity to constrain the meanings of words. Cognitive psychology, 20 (2), 121-57 PMID: 3365937

Merriman WE, & Bowman LL (1989). The mutual exclusivity bias in children's word learning. Monographs of the Society for Research in Child Development, 54 (3-4), 1-132 PMID: 2608077

Merriman WE, Marazita J, & Jarvis LH (1993). Four-year-olds' disambiguation of action and object word reference. Journal of experimental child psychology, 56 (3), 412-30 PMID: 8301246

Healey, E. and Scarabela, B. (2009). Are children willing to accept two labels for one object? Proceedings of the Child Language Seminar. University of Reading.

Systematicity of RNA

2010-05-06T12:00:00.001+01:00

I've been looking at evolutionary precursors to bilingualism. What does this mean? At the moment, I'm thinking about it in the sense of having two or more signals which correspond to the same action or meaning. Not much before language, you say? How about going all the way back to RNA codes?

RNA converts genetic information stored in DNA into proteins which regulate processes within cells. The ‘code’ for translating DNA into proteins is redundant but not ambiguous. There are varieties of code. Different organisms use different proportions of codons. ‘Error’ is defined as sum of protein changes when changing from each codon to each other codon, weighted by the frequency of the codon’s use (Marquez, Smit & Knight, 2005). In this sense, the error rate is comparable with the RegMap index of redundancy.

RegMap was developed to calculate the degree of regularity in the mappings between signals and meanings (Tamariz & Smith, 2008). Essentially, it's the relative entropy modified by the frequency of use.

RegMap was applied to RNA coding frequencies of various organisms. Info was taken from the codon usage database for about 16,500 organisms. As a baseline, the same coding transcriptions were used, but with randomised frequencies. The RegMap index of genetic code and actual usage frequencies is significantly higher than randomised frequencies (Mean RegMap for actual = 0.711, random = 0.708, t = 4.8, df = 7196, p less than 0.0001).

The graph is not much use, but here it is:

Marquez R, Smit S, & Knight R (2005). Do universal codon-usage patterns minimize the effects of mutation and translation error? Genome biology, 6 (11) PMID: 16277746

Monica Tamariz, Andrew D. M. Smith (2008). Quantifying the regularity of the mappings between signals and meanings Proceedings of the 7th Conference on the Evolution of Language. pdf

LEL Postgraduate Conference

2010-05-05T16:00:00.000+01:00

I'm giving a talk at the LEL Postgraduate Conference at the University of Edinburgh, 19th - 21st May. It's not that big of a deal, since I'm required to give a talk, but it is my first talk. A link to the website (which I'm maintaining) with more details and my abstract follow!

LEL Postgraduate Conference 2010

Bilingualism and Social Networks

Children learn language from exposure to speakers in their social network. This learning influences the input that will be given to the next generation. The way languages change over time is dependent on the learning biases of individuals (e.g. Kirby, Dowman & Griffiths, 2007), but also on the dynamics of the social network of those individuals (Gong & Wang, in press; Lupyan & Dale, 2010; Gal, 1979; Govindasamy, 2003).

Bilingualism is often marginalised in theories of language evolution and existing bilingualism is generally seen as the product of contact between two or more monolingual communities. However, I hypothesise that a bilingual ability is a fundamental aspect of language learning: children can learn two languages as easily as learning one. This suggests that human cognition is geared towards handling complex, not homogenous cultural input. This in turn may suggest the kind of social networks in which human cultural transmission evolved. The prevalence of monolingualism in some modern societies may be explained by changes to social structures afforded by communications technology.

This talk will outline my approach to this hypothesis. This involves the idea of cultural transmission as a trade-off between communicative flexibility and expressivity, the use of a comparative approach to bilingualism and methodologies to generate and test hypotheses.

Bilingualism as a preadaptation for Language

2010-05-04T15:56:00.001+01:00

This report is the beginnings of an attempt at a comparative approach to bilingualism, in the style of Fitch(2005). Bilingualism is difficult to define, but by asking whether there is evidence for this capacity in non-human species, it's hoped that this question is made clearer.

FLA and FLB
This research project takes an evolutionary approach to Bilingualism. One of the most difficult problems faced so far is identifying the role of bilingualism in the cultural evolution of language. Is it a product or a catalyst? Firstly, I'm not sure whether this has been considered to any great extent. However, I suggest that the implicit assumption in the vast majority of work in both the areas of Bilingualism and Language Evolution has been that bilingualism is a product of the merging of homogenous language communities. This report explicitly asks the question: Which came first - Language or Bilingualism? That is, did the capacity for bilingualism develop from a pressure to learn multiple existing languages or was it a capacity which existed before human languages were established and influenced their arrival?

The latter hypothesis seems to be non-sensical. How can individuals have the ability to learn more than one language when there are no languages to be learned? Here, I'd like to make a distinction between two kinds of bilingualism, following the approach of Hauser Chomsky Fitch (2002). Bilingualism in the narrow sense means the ability to learn several human languages. This is obviously a human-only trait. Bilingualism in the broad sense refers to the general capacity to acquire more than one signalling system. Depending on how one defines signalling systems, this capacity may be shared with many other animals, both closely and distantly related. Of course, defining what constitutes a single signalling system is difficult, let alone defining language or bilingualism. However, it's hoped that the approach taken in this paper will help towards this goal by considering the features of the phenomenon we wish to define.

Before considering this possibility, the comparative approach to language evolution is presented. Fitch (2005 and others) approaches the study of the evolution of language by considering what elements contribute towards the `Faculty of Language'. In the broad sense of the term, this covers all the prerequisite elements that are required for linguistic communication. This involves cognitive capacities such as acoustic string segmentation and semantic processing, but also much more basic features such as memory. That is, features of the Faculty of Language in the broad sense (FLB) are found in humans and animals. The narrow sense of the term (FLN) refers to those capacities which are involved in language alone. There is much more debate about what these elements are. Recursive processing has been suggested as one example.

The comparative approach has been used to answer the question of what belongs to FLN and to FLB. Animals have been shown to be capable of a number of processes required for language, including categorical perception of speech sounds (Kuhl & Miller, 2978) and Mutual Exclusivity (Juliane & Kaminski, 2004). From studies of divergent and convergent evolution of these traits, some important features have been identified. For example, many species which exhibit vocal learning have direct neural connections between the brain and vocal motors, while non-vocal learners do not (see Doupe, 1999).

This report suggests that this approach should be adopted for the study of Bilingualism. Such an approach would seek to answer whether bilingualism is a uniquely human capacity. If it turns out that other animals also have this capacity, then the role of bilingualism in the evolution of language can be re-assessed.

However, there is a large initial problem. Even FLB only consists of capacities that are required for language. Bilingualism in the broad sense, however, is not required in order to speak a language. This problem may be due to the individual-level bias to the idea of the 'Faculty of Language'. Its primary aim is to describe capacities that an individual organism requires, rather than a community. Therefore, bilingualism may not be part of the FLB, and simply a product of cultural interaction. However, the comparative approach can help verify this hypothesis if social animals exhibit the capacity for bilingualism. That is, if bilingualism comes from cultural interaction alone, there should be no non-social animals which have the capacity for it.

Bilingualism in Bengalese Finches
If other species exhibit bilingualism, then this is evidence that bilingualism developed before human language. Takahasi & Okanoya (2010) study the vocal learning patterns of the Bengalese Finch. These are a domesticated breed descended from wild White Backed Munia. The Bengalese Finch exhibits very complex song patterns in comparison to the White Backed Munia.

Takahasi & Okanoya (2010) carry out a cross-fostering experiment where Munias are brought up by Finches, and Finches are brought up by Munias. The Munias tended to have a stronger preference for copying Munia songs, while the Finches are not so disposed towards their own strain's song. That is, Finches have more flexibility in learning. It is hypothesised that this is because there is a pressure on Munias to identify their own strain in the wild where there are mixed flocks, while this pressure has been masked for Finches by domestication and isolation.

However, is this really `Bilingualism'? The problem is that, although there is flexibility in the sources of acquisition, the birds do not have the same flexibility in production. That is, as I understand it, they still develop only one song (i.e. they can't sing elements of A and B's songs in the morning, then elements of C and D's songs in the evening). Furthermore, the idea of 'comprehension' is more difficult to apply, since there is no semantics.

It has been suggested that Bengalese Finches have developed song complexity as a sexual display (Okanoya, 2004). Following from this, Soma et al. (2009) find that chicks select tutors based on their song complexity. Also, Okanoya (2010) presents some evidence to suggest that Benglaese Finches learn from many tutors. That is, they splice whole segments of songs from many other individuals to create their own song. In this sense, learning from multiple tutors increases the complexity of the song and so increases the attractiveness and fitness of the individual.

The ability to learn syntactic sequences from many tutors has apparently occurred in a system with no semantics. This may suggest that bilingualism at the syntactic level emerged before bilingualism at the lexical level, opposite to the order implicitly assumed by many. One big advantage of cultural evolution is that individuals can inherit information from multiple sources, whereas there are a limited number of biological parents. This is the core of what I mean by Bilingualism being a preadaptation for language: part of the acquisition of human languages requires the flexibility afforded by bilingualism.

Counter Arguments
This phenomenon in Bengalese Finches is interesting, but may not help with our question. Although the complexity of the system has increased due to a change in the environment (domestication), whether this was initially enabled by learning from multiple tutors is not clear.

Also, Okanoya has shown that vocal learners who co-inhabit areas with other species of vocal learners have less complex song. That is, song complexity does not help species identification. Therefore, if the capacity for bilingualism developed in humans before language, it's likely that there was little pressure on vocal cues for species identification.

Inter-species semantic communication
Many species also communicate vocally with other species. Vervet monkeys respond to the territorial and alarm calls of superb starlings (Seyfarth Cheney, 1990). Ring-tailed lemurs respond to the alarm calls of Verreaux's sifakas (Oda Masataka, 1996). However, captive ring-tailed lemurs who had never heard the sifakas' alarm calls also responded appropriately to playbacks. Oda and Masataka argue that they are therefore responding to shared acoustic features rather than to an associated meaning. Although in most examples of inter-species communication do not involve the transference of 'concepts', some examples do show evidence for this.

Zuberbuhler (2000) studied communication between Diana monkeys and Campbell's monkeys. Diana monkeys respond appropriately to Campbell's monkeys' alarm calls for leopards and eagles. Furthermore, their responses suggest they are attending to the meaning rather than the acoustic signal. If a Diana monkey hears a leopard or a leopard alarm call, it calls out loudly, but if it hears a second leopard or leopard alarm, it is quieter, presumably because of the risk of predation (the same is true of eagle alarms). Diana monkeys were primed with Campbell alarms for either leopards or eagles then probed with either eagle or leopard sounds (growls and shrieks). They responded loudly to each combination, apart from where the Campbell alarm corresponded to the predator type (e.g. Campbell leopard alarm followed by a leopard sound). In these cases, the Diana monkeys were quieter, suggesting that they thought the predator was already present.

Zuberbuhler concludes that "Diana monkeys can flexibly use and assess information derived from the communication of other species" and that "semantic understanding can be based on arbitrary signals, as it is the
case for word meaning" (Zuberbuhler, 2000, p. 717). Diana monkeys seem to understand the same concept from two different calls. I argue that this is bilingualism in the broad sense at the lexical-like level.

Again, Diana Monkeys are limited by their physiology in terms of production of the Campbell's alarms. However, the information transfer from Campbell's monkeys to Diana monkeys is not 'communication' as defined by MaynardSmith Harper, 2003) (also see Scott-Phillips, 2008). That is, although the Campbell's calls affect the behaviour of the Diana monkeys, they did not evolve to do this (they are cues, not signals). Therefore, I'd like to suggest that the origins of the capacity for bilingualism originates in the evolution from cues to signals.

However, these responses may not be learned. Furthermore, there is no current evidence to suggest that Campbell's reciprocate in their comprehension of Diana Monkey's calls. The latter issue is discussed by Magrath (2009) who study the alarm call responses of 3 ecologically distinct avian species and find that responses may be reciprocal, but not necessarily symmetrical. Different species reacted to each other's alarm calls in proportion to the 'reliability' of the call as a cue to one of the listener's predators. That is, not all predators of species A are predators of species B, so the A's alarms are not always reliable for species B, and species B responds appropriately. In Magrath (2009)'s study, some species responded in the same way to three different calls. Again, this is evidence for bilingualism in the broad sense.

This raises an interesting question of 'reliability' or 'relevance' (as in Relevance theory, Sperber Wilson, 1995) in animal communications. Much of animal communication is limited to and grounded in information relevant to shared survival interests, that is, food, predators and mating. Humans are capable of communicating about topics beyond their immediate survival needs. This difference possibly requires the of 'ungrounding' of signals from the domains in which they evolved (see the next section).

Bilingualism's impact on FLB
Although bilingualism may not be necessary for the acquisition of language, and so could arguably not be part of FLN, learning two languages does seem to have a qualitative impact on capacities in FLB. For example, compared with monolinguals, bilinguals develop better inhibitory control, theory of mind (Goetz, 2003) and task-switching (Bialystok Martin, 2004).

Raphael Nunez's approach to the evolution of language hypothesises that it involved several pre-adapted 'Modules', but these modules coevolved. That is, an advancement in one module (e.g. more stable voice source, see Demolin, 2010) could cause an advancement in another (e.g. vocal learning), which could feed back into the first module.

Nunez sees the evolution of meaning as involving the development of a grounded system, ungrounding this system from its original domain, then re-grounding it in another. His work focuses on how gestural instantiations of space were re-grounded to convey information about time. For example, one might point behind to indicate an event that occurred in the past. Linguistic expressions of time have also adopted this system.

I suggest that bilingualism can be seen in this way. For instance, being able to learn from several tutors has advantages for increasing signal complexity in some situations. If this ability to learn from individuals could be ungrounded to allow learning from contexts, then this would allow a semantic system to develop. In other words, a kind of bilingualism allows the complex vocal learning mechanisms to be deployed over more general domains.

Okanoya has a similar hypothesis which sees string segmentation and context segmentation as necessary preadaptations for a semantic system. Indeed, the Bengalese Finches studied above may not only be doing string segmentation of tutor's songs, but also a king of crude `tutor segmentation'. That is, they select whole sections from different tutors.

The ungrounding theory suggests that the pressure on the original system needs to be lifted by some other mechanism. This may be a change in the environment, or an internal mechanism. It's likely in the case of the Bengalese Finch that its domestication had a large part to play, alleviating the burden of foraging and predation.

Conclusion
Asking whether non-human species have capacities for bilingualism in the broad sense may affect the way we approach bilingualism. This report has reviewed studies which show that animals have capacities compatible with ideas of bilingualism, but without other features of human language. These capacities stem from very basic evolution of cues and being able to learn from multiple tutors.

Further analysis of evidence for bilingual behaviour in animals is required. These include, for example, switching tasks in primates and other animals and the boundaries between different dialects in whale song. Crucially, this analysis, just like for the rest of FLB, relies on a evidence from a great number of studies. If the relevant studies have not been done, the potential for completing them in this project is extremely restricted.

More fundamentally, this report takes an approach to bilingualism that may not be appropriate. The comparative approach was designed to identify and study necessary components of the language faculty. On the other hand, such an approach may show that, from an evolutionary perspective, there is no easy way to define bilingualism, questioning whether there is a difference between monolingualism and bilingualism or even an easy way to distinguish between languages.

References

Bialystok E, & Martin MM (2004). Attention and inhibition in bilingual children: evidence from the dimensional change card sort task. Developmental science, 7 (3), 325-39 PMID: 15595373

D. Demolin (2010). Prosody and recursion in primate vocalisation Proceed- ings of the JAIST International Seminar on the Emergence and Evolution of Linguistic Communication, Kyoto, Japan.

Doupe AJ, & Kuhl PK (1999). Birdsong and human speech: common themes and mechanisms. Annual review of neuroscience, 22, 567-631 PMID: 10202549

Fitch, W. T. (2005). The evolution of language: A comparative review Biology and Philosophy, 20 (2-3), 193-203 : 10.1007/s10539-005-5597-1

Kaminski J, Call J, & Fischer J (2004). Word learning in a domestic dog: evidence for "fast mapping". Science (New York, N.Y.), 304 (5677), 1682-3 PMID: 15192233

Kuhl, P. & Miller, J. D. (1978). Speech perception by the chinchilla: Identification functions for synthetic VOT stimuli The Journal of the Acoustical Society of America, 63 (3) DOI: 10.1121/1.381770

Magrath, R., Pitcher, B., & Gardner, J. (2009). An avian eavesdropping network: alarm signal reliability and heterospecific response Behavioral Ecology, 20 (4), 745-752 DOI: 10.1093/beheco/arp055

Maynard Smith, J., & Harper, D.G.C. (2003). Animal Signals Oxford University Press, Oxford

Oda, R. and Masataka, N. (1996). Interspecific responses of ring-tailed lemurs to playback of antipredator alarm calls given by Verreaux's sifakas. Ethology, 102, 441-453 : 10.1159/000021651

Okanoya, K. (2004). Song syntax in bengalese ﬁnches: proximate and ultimate
analyses Advance in the Study of Behaviour, 34, 297-346

Okanoya, K. (2010). Biological preadaptations for language Proceedings of the JAIST International Seminar on the Emergence and Evolution of Linguistic Communication, Kyoto, Japan.

SCOTT-PHILLIPS, T. (2008). Defining biological communication Journal of Evolutionary Biology, 21 (2), 387-395 DOI: 10.1111/j.1420-9101.2007.01497.x

SEYFARTH, R., & CHENEY, D. (1990). The assessment by vervet monkeys of their own and another species' alarm calls Animal Behaviour, 40 (4), 754-764 DOI: 10.1016/S0003-3472(05)80704-3

Soma, M., Hiraiwa-Hasegawa, M., & Okanoya, K. (2009). Song-learning strategies in the Bengalese finch: do chicks choose tutors based on song complexity? Animal Behaviour, 78 (5), 1107-1113 DOI: 10.1016/j.anbehav.2009.08.002

Miki Takahasi, & Kazuo Okanoya (2010). Song Learning in Wild and Domesticated Strains of White-Rumped Munia, Lonchura striata, Compared by Cross-Fostering Procedures: Domestication Increases Song Variability by Decreasing Strain-Specific Bias Ethology

Zuberbühler, K. (2000). Interspecies semantic communication in two forest primates Proceedings: Biological Sciences, 267 (1444), 713-718 DOI: 10.1098/rspb.2000.1061

Levels of Bilingualism - update

2010-05-04T15:11:00.000+01:00

I recently went back to my analysis of the Ethnologue into bilingualism. I suggested estimating the level of bilingualism by calculating the minimum number of bilinguals. However, there is already a much better estimation, and it has already been calculated. Greenberg's diversity index calculates the probability of two people who have different mother tongues in the same country meeting. This index has already been calculated for each country in the ethnologue. Below is a map of these indices. Dark colours indicate a higher diversity index (few people have the same mother tongue), lighter colours indicate a lower diversity index (total white = all people have the same mother tongue).

However, this still isn't a good predictor of bilingualism. Instead, it's a measure of diversity of mother tongue. The index still assumes all people only speak one language. I'm trying to figure out how to modify the index to take account of bilinguals - but I was always rubbish at probability.

Keeping Time

2010-04-21T15:47:00.001+01:00

I'm stuck in Amsterdam because of an erupting volcano. We're staying on a barge- very quaint,but the rooms are tiny. I was lying awake, wondering what time it was because my phone has run out of battery. Was it 4am or 12 noon? When should I wake my friends? How to tell how much time was passing? It's not that easy- I remember seeing a competition where people had to judge an hour without watches, and one guy made a bet at half an hour.

It occur to me that singing songs in my head was a good way of keeping time. Instead of counting seconds, I'll count songs. I'll give it two 'beat it's and a 'torn' before waking them, I thought. Then, I thought about the evolution of language conference we visited this week. People were always coming up with theories about the adaptive advantage of language. What if it was useful for measuring waiting time?why would you want to?

In the Miocene, the environment started to dry out, leading to a thinning of resources. This meant that primates either had to reduce their group size or travel further more efficiently in order to find enough food. Our ancestors chose the second option (this theory put forwards by Isbell & Young, 1996).

Now, imagine yourself as part of a large group who travel big distances in forests. Inevitably, you're going to split up. You won't be able to contact them, so you have to decide to wait or move on. In a foreign city with no phone battery, I've been in this situation many times his week. The best hing to do is wait for a while, then move on. But how long? And how to measure?

Singing! And the more complex the song, the less the number of repetitions you have to keep track of. I remember now a piece of child psychology where you tell a child they can have one biscuit now or wait 10 minutes and have two. Intelligent kids will sing to themselves to pass the time.

So there we go, language evolved under an adaptive pressure to accurately measure small periods of time. There are a billion holes in this theory. For example, the sun is a pretty good indication of the time. Also, it's not clear that this ability is any use.

Anyway, we might get a few papers, a book and a conference out of it.

Isbell, L. A., & Young T. P. (1996). The evolution of bipedalism in hominids and reduced group size in chimpanzees: alternative responses to decreasing resource availability Journal of Human Evolution, 30 (5), 389-397 DOI: 10.1006/jhev.1996.0034

Catch Up

2010-04-12T10:24:00.001+01:00

This weekend I took part in a 48 film making competition! Here's the result:
http://www.dailymotion.com/video/xcwyzb_catch-up_shortfilms

Yuna

2010-04-05T23:11:00.000+01:00

I've just finished a stencil for a friend. It's Yuna from Final Fantasy X. I used too much paint on the first print, but the second came out much sharper.

Step 1: Steal picture

Step 2: Print and Cut

Step 3: Spray

Step 4: Repeat

Cultural Variation and Social Networks

2010-04-01T15:40:00.008+01:00

Children learn language from exposure to speakers in their social network. This learning influences the input that will be given to the next generation. The learning biases that an individual has will influence the way the language changes over generations (Kirby, Dowman & Griffiths, 2007). However, language also plays a part in constructing and maintaining social networks. Recent studies have suggested that the structure of the social network also has an effect on the how a language evolves. Gong and Wang (2010) find that different network types influence the evolution of linguistic categories in an artificial categorisation game. Lupyan & Dale (2010) find that the amount of contact with other communities, and a community's spatial dispersion influences the morphological complexity of a language.

I was wondering whether bilingual communities have different social network structures to monolingual communities. Real social networks are very difficult to construct, so I wanted to use some online social networking sites. Twitter seemed like an obvious choice because of it's simple API, and also because 'following' someone has a genuine connection on a user's linguistic input.

I aquired some data for Twitter users. The data includes the number of followers (indegree) and people being followed (outdegree), the user's location, the number of status updates sent by a user and the amount of time since the last update. The last two features can be used to filter out people who are not active participants. The location information is optional and may be as specific as GPS coordinates or as general as a country, or even just a timezone. Following from this, communities were defined by country. Data mining techniques will be used to automatically assign users to countries.

Ideally, we would want the following statistics: Average Degree, Clustering Coefficient, Average shortest Path length. However, this requires information on the specific links between users. However, this requires more time and resources, so this data was not collected for this report. This is not a trivial point, however, because users can follow people in other countries.

Next, data on the linguistic variance is needed. As I showed in a recent post, estimating the amount of bilingualism is difficult. The best source of information is Ethnologue, but numbers of speakers are underestimated, probably due to inadequate data for small linguistic communities. I decided to use two measures of bilingualism: The number of languages spoken in a country and the percentage of the population of a country that speak the majority language. A country's nominal per-capita GDP and number of internet users is also taken into account.

Data was collected for about 31,000 users, about 25,000 of which was usable (there have been databases of up to 2.7 million users with over a billion connections between them). Although Twitter allows very many followers, for practical purposes I filtered out users with over 1,000 followers or following over 1,000 other users. This left data for 17,444 users ffrom 119 countries.

Initial results suggest a negative correlation between the indegree for each country and the percentage of the country's population who speak the majority language (using log indegree, t = 1.88, df = 117, p = 0.06).

Using a linear regression, the indegree and outdegree are significant predictors of linguistic variation, even when the effects of population size and access to the internet are partialled out (R-squared = 0.19, F(4,17439) = 1063, p <0.01; t =" -2.04," p =" 0.04;" t ="2.54," p ="0.01). This was based on data for 17,444 users ffrom 119 countries. Statistics for countries were taken from CIA factbook, 2010. The analysis revealed a negative correlation between linguistic variance and indegree, but a positive correlation between linguistic variance and outdegree. The same qualitative results were found by using the number of languages spoken in a country. However, there is a positive correlation between the number of languages and both indegree and outdegree. I'm not sure how to interpret this yet, or whether any of it makes any sense. In the meantime, here's a pretty uninformative map of the world, coloured by average number of Twitter friends. Darker countries have users with a higher average number of friends.

Lupyan G, & Dale R (2010). Language structure is partly determined by social structure. PloS one, 5 (1) PMID: 20098492
Kirby S, Dowman M, & Griffiths TL (2007). Innateness and culture in the evolution of language. Proceedings of the National Academy of Sciences of the United States of America, 104 (12), 5241-5 PMID: 17360393

MSGs and FSGs

2010-03-14T13:12:00.001Z

I'm writing this sitting at the bar of a busy Chinese restaurant. I can see the chef work at almost miraculous speeds, divinding his time between tossing the wok, fetching ingredients, mixing and adding the sauces, then serving, cleaning and starting again. The whole cycle is very complicated and happens in about two minutes. The chef must have essentially written some of the routine to muscle memory, just as I did when I was working as a pizza chef. Then I notice something that brings back a hint of nostalgia for the greasy pizza kitchen. The chef, after a complicated movement scooping sauce from a bowl at neck height, transferring it to the wok, switching ladels and tossing the wok, he strikes the counter with the ladel. As far as I could tell, this had no purpose such as clearing food from the ladel. But I used to do it too, just after I finished cutting all the pizzas in an order. Unconsciously, I would strike the work surface with my tongs. Indeed, the chef in front of me turns away to begin a new order as soon as he completes his coda.
The reason I'm writing about this is that I've just been to a lecture by prof. Okanoya, the famous academic studying the bengalese finch- one of the singbirds with the most complicated song syntax in the world. Okanoya has generated a finite state grammar which captures these songs, and, interestingly, most of them include a sequence that only appears at the end of a phrase.
Basically, I was wondering if the chef and I had built our own cullinary finite state syntax, and whether we instincively added our own 'end note' to help conordinate ourselves. We may share a lot more with animals in our cognition than we might suspect.

The Language Evolution Tree Returns

2010-02-17T18:35:00.000Z

Recently, I found that people have an innate bias to associate Evolutionary Linguistics with Acacia trees. I've just noticed that Babel's Dawn, a blog about the evolution of language by Edmund Blair Bolles also has an acacia tree in the sunrise on its banner. Alas, it's not the language evolution tree, but it's pretty close. With this in mind, I propose that there is a deep association between acacia trees and language evolution. Just look at the following two graphs ...

Distribution of Acacia Trees

Distribution of Languages using complex tone systems

Coincidence?
(yes, but this isn't)

How many words for Red? Part 4

2010-02-16T16:27:00.003Z

Xan Gregg has done a data quality analysis of the Wikipedia colour data I used in recent posts (here). Unsurprisingly, the data is not great quality. The outliers discussed are not included in my analysis, but the colour conversions are not totally consistent either. As Gregg points out, Wikipedia is hardly a good source for Visual Psychophysics research, but it's still an interesting proof-of-concept.

How many words for Red? Part 3

2010-02-11T17:04:00.001Z

This week I've been writing about a bin-packing approach to the bilingual lexicon (here and here). Here's why:

The mutual exclusivity bias is a default approach infants have to learning words which assumes that every object has a single, unique name. However, some studies show that bilingual infants do not follow this bias. The question I'm researching is why assume different things about the world if you're hearing two languages instead of one? Certainly, bilingual children hear more synonyms, but monolinguals also hear many words for the same objects.

The literature has not produced clear-cut results (see my posts here and here), so I've been using a simple model to try and organise my ideas. The model is based on the Categorisation game, where a population of agents try to agree on words for colours. That is, a speaker is presented with a scene of several colours and refers to one of them. A listener must decide which colour the speaker is talking about. Agents begin with no categories and no words, but then divide the colour spectrum into categories and associate words with them, based on verbal interactions. The algorithm is reproduced here.

The algorithm makes some assumptions about mutual exclusivity. The first (what I call Heuristic A) is that, when you see two objects within the same category (e.g. two shades of red), you should divide that category so that there is only one object in each, then assign to each a new unique name. That is, assume that different objects have different names. The second (Heuristic B) is that, when you communicate successfully, delete all other names associated with the category.

These two heuristics limit bilingualism and introduce a mutual exclusivity bias. I ran the categorisation game model without these heuristics to see what would happen. Below are the results of two runs - one with Heuristic A (Black), and one without (Red) (10 runs each, 25 agents, a maximum of 100 perceptual categories, 20,000 rounds). Measures include the number of perceptual categories (both rise and plateau at the same rate), communicative success rate (both similar), the average number of names an agent has, the bin packing depth (bpDepth), bin packing wastage (bpE) and the amount of lexical overlap (overlap function from Baronchelli, Gong, Puglisi and Loreto, 2010).

Removing this heuristic has some interesting consequences for the model. Firstly, the average communicative success is unaffected by removing heuristic A. The number of perceptual categories also increases to the maximum in the same timescale. However, removing the heuristic leads to agents which are more memory-efficient in terms of the number of labels they know, and the efficiency of those names to describe the meaning space. That is, the bin packing metric suggests fewer synonyms and a more efficient coverage of the meaning space. In fact, agents without heuristic A were near optimal in their bin packing.

I suggest that removing heuristic A (each different object has a different name) changes the demands on memory in such a way to favour agents that have several complete descriptions of the meaning space. Dropping heuristic A also reduces the number of lexical items that are stored.

Children exposed to two languages have extra demands on lexical memory. The above analysis suggests that it's a good idea for these children to drop heuristic A in order to save storage space. That is, if you're bilingual, you shouldn't assume that every object has a different name. Indeed, this predicts some of the findings in the experimental literature (i.e. that bilinguals do not apply mutual exclusivity).

However, part of the problem is that this model is a model of emergent structure in labelling perception, it's not a model of acquisition. The method of splitting a perceptual space into categories is also possibly not realistic.

I've just come back from a talk by Kenny Smith, who's been running experiments into Mutual Exclusivity. He trained participants to associate novel words with novel objects, with some participants getting more synonymy than others (objects may have two associated words). After this, participants did a mutual exclusivity task - they were shown two objects, one from the training set and one new object and were asked which one was associated with a novel word. The degree that the participants adhered to mutual exclusivity was proportional to the amount of synonymy they had experienced.

Therefore, it seems that deciding to drop the mutual exclusivity bias may occur on-line. It remains to be seen whether the same results are obtained for children.

The Minimal Naming Game

2010-02-10T13:44:00.002Z

This is the Minimal Naming Game Algorithm from Puglisi et al. (2008). I point out two Heuristics that affect the agent's ability to acquire two langauge systems.

There is a population of agents, each with a partitioning of the perceptual space called categories. Each category has a list of associated words. Each agent has a minimum perceptual difference threshold dmin , below which stimuli appear the same. At each time step:

1. Two individuals are chosen at random to be the speaker and the listener.

2. They both have access to a scene containing M stimuli. The stimuli must
be perceptually distinguishable by the agents (perceptual distance ≤ dmin ).

3. The speaker selects a topic and discriminates it in the following way:
• Each stimulus is assigned to a perceptual category
• If one or more other stimuli are assigned to the same category as the topic, the agent splits its perceptual categories so that each stimulus belongs to only one perceptual category.
• The new partitions inherit the associated words of the old partition.
• Heuristic A: Each new partition is given a new, unique name.

4. The speaker transmits a word that it associates with the topic to the listener.
If it has no words associated with the category, it creates a new one. If it has more than one word associated, it transmits the one that was last used in a successful communication.

5. The hearer receives the word and ﬁnds all categories which have the associated word and which identify one of the stimuli in the scene. Then:
• If there are no such categories, the agent does nothing.
• If there is one such category, the agent points to the associated stimulus.
• If there is more than one such category, the agent points randomly at an associated stimulus.

6. The hearer discriminates the scene, as above.

7. The speaker reveals the topic to the listener.

8. If the hearer did not point to the topic, the communication is a failure. The hearer adds the transmitted word to the category discriminating the topic.

9. If the hearer pointed to the topic, the communication is a success.
Heuristic B: Both agents delete all other words but the transmitted one from the inventory of the category discriminating the topic.

How many words for Red? Part 2

2010-02-09T11:29:00.007Z

Just an update on the last post (here). I looked at the distribution of colour terms from Wikipedia according to their hue value and suggested that people have more words for some colour ranges than others.

First, I ran a linear mixed effects model on the data, and came up with slightly different results. The spectrum was split into 10 equally sized bins, and the number of words that fell into each bin was counted. This number was predicted by the bin number and the slope and intercept were allowed to vary by language. The bin number significantly improved the fit of the model (Log likelihood difference = 5.12, Chi squared = 10.2, df = 1, p=0.001). This suggests that the distribution is not flat (i.e. there are more words for some colour ranges than others). However, allowing different languages to have their own fit did not significantly improve the model (Log likelihood difference = 0.65, Chi squared = 1.3, df = 2, p = 0.52). This suggests that languages do not differ significantly in the distribution of colour words.

However, I also noticed that the distribution looks very similar to the Just Noticeable Difference (JND) curve for colour. Human eyes are not uniformly sensitive to colour. We can distinguish colours better at some ranges than others. Below is the distribution for colour terms from the English Wikipedia site, with an overlay of the human JND curve (from Long et al., 2006).

You'll notice that the curve is a very good fit. Indeed, the two distributions are correlated (r=0.6, df=18, p=0.005). That is, the distribution of colour words may not be uniform over the physical spectrum, but it is pretty even across the perceptual spectrum. Put another way, humans have lots of words for ranges of the spectrum that they are good at discerning.

For 8 languages, the number of colour categories and the JND are correlated (r = 0.28, df = 94, p = 0.005), and more so for all non-monochromatic colours (r = 0.3, df = 93, p = 0.002814).

A mixed effects model shows that the perceptually normalised number of colours (num
colours/JND) are still significantly skewed (Log Likelihood difference = 10.22 Chi square = 20.4234 p< 0.001). But this skew is not much different between languages (Log Likelihood difference = 1.95 chi square = 3.8, p= 0.14). (The p values drop to 0.0002 and 0.55 when considering non-monochromatic colours)

This suggests that there is still a non-perceptually motivated colour term distribution
skew.

Long F, Yang Z, & Purves D (2006). Spectral statistics in natural scenes predict hue, saturation, and brightness. Proceedings of the National Academy of Sciences of the United States of America, 103 (15), 6013-8 PMID: 16595630

How many words for Red?

2010-02-08T16:06:00.004Z

Just how different are languages in the way they label colours? Since Berlin & Kay's 1969 study of colour terms in many languages, the debate over cross-linguistic similarities has raged. Interestingly, most of the subjects in Berlin & Kay's experiments were bilingual, but they didn't think that other languages could influence the results of individuals.

In the last few years, Dr Panos Athanasopoulos at the ESRC Centre for Research on Bilingualism, Bangor, Wales, has been investigating colour perception in Bilinguals. In 2009, Dr. Athanasopoulos studied Greek-English bilinguals. Greek makes a distinction between dark blue ble and light blue ghalazio. Results suggested that bilinguals' perceptions shift towards those of native speakers of their second language. The study is set to be extended into Japanese this year by two forthcoming publications (in Bilingualism: Language and Cognition and Language and Bilingual Cognition).

Last week I was thinking about models of colour terms and bilinguals. What do we mean when we say someone is 'bilingual'? On a syntactic level, this may be a bit easier to answer, but on the lexical level (where I am at the moment) is a bit more difficult. For example, what's the difference between an having two 'languages', and having one 'language', but many words for the same category?

My current suggestion is that bilinguals differ from monolinguals because they have more sets of categories that span their entire perceptual field. What does that mean? Imagine the colour spectrum (red -> yellow -> green -> blue). A monolingual will be able to label any point in the spectrum with one word, and some points with more than one. For instance, a certain red may be 'red' and 'crimson'. However, a bilingual will be able to label any point with more than one word. That is, the spread of their synonyms will be more even.

Here's an illustration: Below is how a monolingual might break up the colour spectrum. It has several categories in its memory (the colour 'E' spans from red to yellow, 'C' covers greens and 'A' covers a specific kind of green, etc.):

The bottom of the image is labelled 'bin packing'. This represents the results of the bin packing algorithm, which tries to fit the categories into the smallest space possible. The monolingual above has lots of wasted space (striped areas) because it has lots of synonyms for a few categories.

The next illustration is of a bilingual, but note I haven't specified which 'language' a category belongs to. They have synonyms too, but they pack much more efficiently. In fact the speaker below has two options with which to describe any colour. In contrast, the monolingual has only one complete system.

Ok, these examples are set up. But I still predict that bin packing of bilingual categories will be more efficient than for monolingual categories. I'll use this metric to examine the results of a model in an upcoming post.

However, this metric assumes that colour categories within a language are unevenly dispersed. That is, speakers know many words for some colours, but not others. Let's put this to the test. The problem with doing so is that most colour category experiments involve getting people to assign colours to labels, meaning that they can't declare two labels for the same colour. So, I went to the Wikipedia List of Colours page which lists details of all colours mentioned in technical articles on colour. These are mainly standardised names for use in HTML, which is a problem for our current analysis, but let's see what happens.

The names for colours in English and other langauges were gathered along with the Hue angle (i.e. position in the spectrum). Below is the histogram of the number of colour terms in different portions of the spectrum (coloured by average colour of the colour terms, taking saturation and brightness into account) for English, followed by the histograms for some other languages:

First, the histogram for English is certainly not even. There are more names for reds and blues than for greens. This works intuitively - how many types of red can you think of in comparison to types of green?

Let's look cross-linguistically: First of all, the histograms are not identical - the Wikipedia pages are not just translations. Second, they all seem to have lots of names for reds. This may be an artefact of the circular meaning space (from 0 degrees to 360 degrees), but would not entierly explain the imbalances. Let's do an ANOVA of the number of colour terms per section of the spectrum (splitting into 10 even sections) by section position and language.

Colour names were not evenly distributed within languages (F(1,64)=11.35, p<0.01), but were significantly differently distributed across languages (F(7,64)=6.31, p<0.001). Having said this, I'm not completely sure of my stats here.

However, if the analysis is correct, then bilinguals should have a better packing efficiency than monolinguals because acquiring a whole extra 'language' is more likely to normalise the distribution rather than increase the skewness. This may be a useful metric in the analysis of feature-level models of bilingualism.

Linguistic Leniancy

2010-02-05T14:49:00.000Z

I was once on holiday with a motley crew of tourists when a guy called Dave decided to show off his linguistic skills. He'd lived in China for a few months, and picked up some Mandarin, which he proceeded to flaunt in front two girls from Hong-Kong. However, no matter how hard he tried, the girls had no idea what Dave was saying. After he explained in English what he was trying to convey, the girls did recognise the traces of meaning. He was partly faulty on tone (notoriously difficult for foreigners) and partly on pronunciation.

Dave was very disappointed and a bit embarrassed, but also confused. How had he managed to order all those meals, to greet all those people, to have all those conversations, if he couldn't actually be understood? One possible answer is, instead of Dave learning Chinese, the Chinese had learned Dave. That is, after a few interactions and a lot of contextual help, they had understood that a particular incoherent sound meant that he wanted some noodles/was asking the time/was saying hello.

Recently I've been thinking about what happens when two populations with different languages mix. Several studies have considered this using computational models. Several suggest that one language usually takes over, and that bilingualism is not stable. What the Dave's example suggests is that speakers are extremely forgiving when dealing with people who do not speak their language, especially when gesture and context can make most meanings clear. It's also a good example of the fact that people can communicate despite large discrepancies in their mental representations.

Pun of the Day #7

2010-02-02T18:27:00.002Z

L: I need more washing powder... did I say that already?
Me: Yes, you're experiencing déshàmpoo.

The Language Evolution Tree

2010-01-27T13:24:00.001Z

According to one theory of language evolution, structure emerged gradually, with sub-parts of words being re-interpreted by learners as referring to sub-parts of the things they referred to. However, for this process to work, there must be some chance similarities between words and referents to begin with. This has been a major criticism of the theory, but here I present some evidence that such chance coincidences do occur (tipped off by a colleague).

Below are the front covers of two books on language evolution. The first is Christiansen & Kirby's Language Evolution, the second is Tecumseh Fitch's recent publication, The Evolution of Language.

The first thing you'll notice is that they both feature trees in twilight. Strange enough perhaps, especially when it's not really that obvious what trees have to do with language evolution. However, a further analysis reveals an even greater coincidence:

It appears as though both covers feature the same tree. Indeed, looking up the images revealed that both are property of CORBIS, and both were taken in the Massai Mara Game Reserve, Kenya.

From now on, Language Evolution and Acacia trees will be forever linked because of a chance occurance. There are even rumours of a trip to visit the 'Language Evolution Tree'.

And so structure emerges from chaos ...

Evolutionary Linguists announce arrival of Skynet

2010-01-26T09:00:00.000Z

In a recent book chapter, Loreto, Baronchelli & Puglisi (2010) summarise mathematical models of language games. They minimally define a system which will allow agents to reach shared linguistic categories which describe continuous stimuli (e.g. colour). New findings include a demonstration that the number of linguistic categories will stay low, regardless of the resolution of the agents' 'eyes'. However, I was particularly struck by the last paragraph, which appears to be ominously more 60s science-fiction than the mathematical jargon:

"Finally it is important to mention that in the last few years a potentially very interesting experimental platform appeared: the World Wide Web. Though only a few years old, the growth of the Web and its effect on the society have been astonishing ..."

What is this wondrous thing called the World Wide Web? Why has nobody told me about it?

Yes, as far as I can tell, this was published last month. Then, the really spooky Skynet stuff:

"Innovation has widened the possibilities for communication. Social media like blogs, wikis, and social bookmarking tools allow the immediacy of conversation, with unprecedented levels of communication speed and community size. In this perspective the web is acquiring the status of a platform for social computing, able to coordinate and exploit the cognitive abilities of the users for a given task."

We are enslaved by this terrible self-organising system! But what task is it trying to complete? Surely nothing good. It seems that Evolutionary Linguists just can't help imagining dystopian futures.

Vittorio Loreto;, Andrea Baronchelli;, & Andrea Puglisi (2009). Mathematical Modeling of Language Games Evolution of Communication and Language in Embodied Agents, 263-281

Language Structure and Social Structure

2010-01-25T13:21:00.001Z

Last week saw the publication of Lupyan & Dale (2010) (also discussed here). It's an analysis of languages from the World Atlas of Langauge Structures (WALS, quite fun to play around with), showing that languages spoken by more people tend to be less morphologically complex (fewer cases, fewer inflections, more of a tenancy to express things using separate words rather than with morphology). It's hypothesised that this is because a greater, more dispersed population will contain more second-language learners, therefore the language will tend to change to be easier to learn by adult learners, who seem to find morphology more difficult to learn than child learners.

The number of speakers in a language certainly varies a great deal. Lupyan & Dale emphasise this by pointing out that the median number of speakers in a language is 7,000 while the mean is 828,000. The number of second-language learners is also not a trivial influence in modern times. For example, only about 30% of English speakers are native speakers. Even more extreme is Malay with only 15% of speakers learning it as a first language.

In comparison, Siberian Yupik Eskimo has essentially no non-native speakers (from supporting data). In such communities, there is a greater pressure for the language to be learnable by children, therefore it retains morphological complexity.

The changes introduced by adults are likely to be acquired by children. Thus, Lupyan and Dale talk about exoteric and esoteric languages. The changes introduced by adults are likely to be acquired by children. Although the paper also takes into consideration the fact that children do not always learn from their parents:

It has been argued that there is no automatic transmission of the “mother tongue” from parents to offspring (47). For example, in a survey of 188 individuals in Senegal who listed Bambara as their native language, Bambara was the father’s native language in 16%, the mother’s in 19%, the native language of both parents in 26%, and the native language of neither parent in 39% (47).

Interestingly (for me at least), Welsh is pointed out in the regression graph and seems to fit the pattern - it has relatively few speakers and a high complexity score (calculated by "summing the number of features for which each language relies on lexical versus morphological coding and subtracting the total from zero"). While Welsh speakers were dispersed throughout Britain in the 8th century (and a Welsh speaking colony remains in Argentina), there are probably more child speakers than adult speakers currently. However, contact with English has introduced a bias for lexical forms over morphological forms (at least on my schoolyard - consider "Rydw i'n ysgrifennu" vs. "Ysgrifennaf"). Coupled with a pressure from language conservation groups to make Welsh more accessible to learners (see my post here), Welsh may indeed become less morphologically complicated.

Also interesting is the comparison with language data from Ethnologue, noting that the WALS seems to under-represent sub-Saharan Arfican languages (here). It appears that the Ethnologue is still the best source of data on population sizes, even though it appears not to be good enough to calculate the number of bilingual speakers (here).

One question is to what extent this phenomenon of exocentrism is a modern one. Large social groups and extensive travel only really took off in the last thousand years, so it's not clear if these dynamics can be scaled backwards in time to aid study of the evolution of language. It's been argued that human evolution allowed larger group sizes in comparison to chimpanzees (Isbell & Young, 1996), did this force communication to become more structured? Was language primarily lexically-based and used by adults?

Perhaps, however, it could be applied to animals - do songbirds living in larger, migratory flocks have simpler song morphology than those in smaller, more localised ones? The complexity of the domesticated finch is certainly more complex than its wild descendants (Honda & Okanoya, 1999). Could simply reducing the number of 'speakers' be an influence?

It's a fascinating article, and has already stimulated a lot of debate.

Lupyan G, Dale R (2010). Language Structure Is Partly Determined by Social Structure PLoS ONE, 5 (1) : 10.1371/journal.pone.0008559

Honda, E., & Okanoya, K. (1999). Acoustical and Syntactical Comparisons between Songs of the White-backed Munia (Lonchura striata) and Its Domesticated Strain, the Bengalese Finch (Lonchura striata var. domestica) Zoological Science, 16 (2), 319-326 DOI: 10.2108/zsj.16.319

Isbell, L. & Young (1996). The evolution of bipedalism in hominids and reduced group size in chimpanzees: alternative responses to decreasing resource availability Journal of Human Evolution, 30 (5), 389-397 DOI: 10.1006/jhev.1996.0034

Cymraeg Clir

2010-01-21T14:13:00.001Z

A scheme at Bangor University offers translation services in Welsh. But this one doesn't translate into English. Instead your documents come back in Cymraeg Clir (Clear Welsh). Ok, so it's more of an editorial service really, but there are some interesting features in its guidelines (The document is titled "Ysgrifennu'n Glir" (Writing Clearly), even though it recommends using 'sgrifenny' instead of 'ysgrifennu'):

Move the subject closer to the start of the sentence by using active verbs, rather than passive. This is general advice for any language, but there are Welsh-specific ones too:

Use verbs instead of nouns. This is slightly confusing at first glance, but refers to the use of phrases like 'tree management' rather than the simpler 'managing trees'. The incursion of such nouns is put down to the influence of English.

Use the 'long form' of the verb (uninflected) to avoid tense ambiguity.

Interestingly, the use of acronyms is discouraged, partly for ease of reading but also because acronyms can be misleading, or even difficult to construct, due to initial consonant mutation (the changing of the first consonant in a word in some syntactic contexts, e.g. 'cat' in welsh is cath, but 'his cat' is ei gath). However, established acronyms and English acronyms along with full Welsh translations can be used. Furthermore, acronyms themselves cannot be mutated.

Mutation is also to be avoided in bullet points by inserting e.g. 'the following' before them (all contiguous nouns after a mutation context are elligable for mutation). Interestingly, foreign place names should not be mutated either.

Users are reminded that mutation applies by agreement, not just in immediate contexts, so in the following sentence, since 'neges' is a feminine noun, the verb 'teipio' (to type) must be mutated using 'aspirate' mutation:

Dyma’r neges y mae’n rhaid i mi ei theipio
Here the message it is must to me it type
This is the message that I must type

Whenever rules like this need to be explained in editorial documents, I wonder whether Welsh mutation has just gone too far.

Contracting the verb 'is' (e.g. 'the building's insured') is recommended. However, the Welsh 'yn' (is) can also mean 'within', in which case it should not be contracted. Instead of explaining the secondary meaning in Welsh, however, the guideline uses English:

"heblaw pan fo’r ‘yn’ yn golygu in yn y Saesneg"
(exept when the 'in' means in in English)

Never use the ampersand (&) when writing in Welsh! This makes sense, since it's a contraction of a Latin phrase anyway.

There is also advice on punctuation, including the use of hyphens to separate double letters, avoiding the possibility of triple and quadruple ls (e.g. alllifo).

Ghost in the Shell Tattoos

2010-01-13T15:07:00.001Z

Following my obsession with Masamune Shirow's Ghost in the Shell, here are some GITS-inspired tattoos. I also found one of the laughing man icon, but couldn't link it. Surprisingly, I couldn't find any individual eleven tattoos, but here's someone who's planning to get one.

By Stanio

By Kryoide

By tetsuonabiki

By Scott White

A Tachikoma from Thursday Morning

By Melissa

By Craig B.

By rkampschoer

By Neon-Stiches

By Todo

From Riokou

By Romantist-Egoist

From M1K3Y

Evolutionary linguists employed by Aliens

2010-01-12T09:19:00.000Z

As I slowly weave my way deeper into the underworld of Evolutionary Linguistics, I can't help thinking there's an alien conspiracy going on ...

Walking around the basements of Edinburgh Universtiy, signs can be seen inviting you to 'Participate in an Alien Language Experiment'. Indeed, Kirby, Cornish and Smith (2008) tested whether humans could "learn an alien language". In a recent Horizon program, Prof. Kirby can be seen extending his encounters of the third kind in what appears to be a bombed-out Victorian Prison - clearly a secret base for some dark organistation:

Gary Lupyan, recently seen giving a talk in Edinburgh, has run experiments which test whether humans can tell the difference between "approachable" and "dangerous aliens.

Further, in Ichinco, Frank and Saxe (2009)'s study, "participants were told that they had been kidnapped by aliens who were trying to teach them their language through two episodes of alien television". I was under the impression no signals had been received from other planets, never mind double-features.

It also emerged that Evolutionary Linguist Gareth Roberts, had been asking participants to imagine themselves competing for resources on an alien planet.

From this, I can only deduce one thing: Aliens have been employing evolutionary linguists to decide which humans to kidnap and bring back to their planet as slaves. So, the next time you find yourself asked to imagine that you're learning how to mine heavy metals on a distant planet or quizzed about whether you would be likely to oppose oppressors with green tentacles, feign incompetence.

Summary: Grassman and Tomasello (2010)

2010-01-11T14:57:00.000Z

Grassman and Tomasello recently published a study in reaction to Jaswal & Hansen (2006). Jaswal & Hansen found that children ignore pragmatic cues (pointing and looking at objects) in favour of honouring mutual exclusivity. In the first experiment, 24 children (3-4 y.o) were presented with a novel and a familiar item (see my post on diagrams). They were split into two conditions where the experimenter pointed at the familiar object and either asked 'Can you hand me the X?', where X was a novel noun, or made a neutral request ('Can you give it to me?').

All children in the latter 'pragmatic baseline' condition selected the familiar object the experimenter was pointing or looking at, while almost all children selected the novel object when presented with a novel noun. In a second experiment, a similar setup was used, but where the experimenter shifted their gaze towards one object instead of pointing and used an arbitrary label which could refer to any of the objects (e.g. 'Can you give me the one I got yesterday?') instead of the neutral . Here, children presented with a novel label chose the familiar item significantly less often than those in the pragmatic baseline.

The children were aware of the pragmatic cues: Children in both conditions monitored the experimenter's cue equally often. Furthermore, Jaswal & Hansen report that some children, when asked for a novel noun and presented with a pragmatic cue (looking at the familiar object), checked under the table on the side suggested by the pragmatic cue then selected the novel object anyway.

However, Grassman & Tomasello (2010) question Jaswal and Hansen's conclusions, pointing out that gestural cues are ontogenetically and phylogenetically primary to conventional, linguistic cues.

Furthermore, they argue that Jaswal & Hansen's pragmatic cues were non-ostensive. That is, the experimenter's gaze did not follow the pointing or, in the looking condition, there was no gaze alternation. Thus, there was no cue that the pointing/looking was a communicative act intended for the child. Furthermore, they used 'small' pointing, which is associated with redundant information rather than instead of 'big' pointing which is associated with primary information.

Grassmann & Tomasello extend Jaswal & Hansen's experiments with direct linguistic cues ('Give me the X', where X was a novel noun) instead of indirect ('Give it to me'). Two and four year olds showed no preference for linguistic over non-ostensive cues. However, without pragmatic cues, children chose the the novel object, while with ostensive cues (pointing with whole arm, gaze alternation) children chose the familiar object (see my post on diagrams). These directly oppose Jaswal and Hansen's results.

In a second experiment, children were presented with a familiar and novel object and requested a the familiar object verbally while pragmatically cuing the novel item. Children chose the novel object, following the pragmatic cue.

In a third experiment, the experimenter presented two familiar objects while verbally requesting one and pragmatically cuing another. Here, there were no preferences for following the pragmatic or lexical cue. 2 year olds exhibited mixed responses across trials, while 4 year olds were consistent, but with half following the pragmatic cue and half following the lexical cue.

Grassmann & Tomasello place this data in the context of the child's conflict resolution and lexical entries. Children try to integrate cues from different domains into one coherent communicative intention. It is suggested that it may be harder to modify lexical entries for familiar words without a clear reason than to link novel words to familiar objects.

Jaswal VK, & Hansen MB (2006). Learning words: children disregard some pragmatic information that conflicts with mutual exclusivity. Developmental science, 9 (2), 158-65 PMID: 16472316

Grassmann, S., & Tomasello, M. (2010). Young children follow pointing over words in interpreting acts of reference Developmental Science, 13 (1), 252-263 DOI: 10.1111/j.1467-7687.2009.00871.x

Mutual Exlcusivity Diagrams

2010-01-11T14:26:00.006Z

Lately, I've been reading about experiments on Mutual Exclusivity. I've found it useful to draw up some diagrams to illustrate each. I've posted some of them below. Each diagram represents entities in the Singal, Concept and Environment domains. Circles represent entities (signals, concepts or actual objects). Lines between circles represent a link between them. For instance the word "Dog" would be linked with a concept "Dog" which could be linked to an actual instantiation of a dog in the environment (as long as the observer had learnt these). Dotted circles indicate a novel entity which the observer has not seen before. Dashed lines indicate links trained by the experimenter.

Au & Glusman (1990):
Byers-Heinlein & Werker (2009):

Jaswal & Hansen (2006):

Grassman & Tomasello (2010) - a reaction to Jaswal & Hansen above:

Healey & Scarabela (2009):

Au TK, & Glusman M (1990). The principle of mutual exclusivity in word learning: to honor or not to honor? Child development, 61 (5), 1474-90 PMID: 2245739

Byers-Heinlein, K., & Werker, J. (2009). Monolingual, bilingual, trilingual: infants' language experience influences the development of a word-learning heuristic Developmental Science, 12 (5), 815-823 DOI: 10.1111/j.1467-7687.2009.00902.x

Jaswal VK, & Hansen MB (2006). Learning words: children disregard some pragmatic information that conflicts with mutual exclusivity. Developmental science, 9 (2), 158-65 PMID: 16472316

Grassmann, S., & Tomasello, M. (2010). Young children follow pointing over words in interpreting acts of reference Developmental Science, 13 (1), 252-263 DOI: 10.1111/j.1467-7687.2009.00871.x

Healey, E. and Scarabela, B. (2009). Are children willing to accept two labels for one object? Proceedings of the Child Language Seminar. University of Reading.

Pun of the Day #6

2010-01-11T14:24:00.001Z

Trying to get past J with plate of chips

J: I'm afraid to get past me there's a chip tax.
Me: A potatoll?

Synaptic Gasp

2010-01-09T14:33:00.000Z

A while ago, I posted my own stencil inspired by Ghost in the shell. Here's Synaptic Gasp by ocean.flynn which has similar themes:

ocean.flynn says:

The synaptic cleft in the human brain reminds me of the gap between the hand of God and Adam in Michaelangelo’s visualization of Creation. My mind is stuck on the image of the gap. That’s the leap of faith between that which we can know and that which is beyond our capacity to know.

Inferential Models of Bilingualism

2010-01-06T15:02:00.001Z

de Boer (2000)'s models showed that shared representations in a communicative system could be achieved given certain principles. Steels & Belpaeme (2005) extended this framework to the domain of colour, which is a bit more intuitive to understand: A speaker and a hearer are placed in an environment full of coloured tiles. A speaker chooses a colour and names it (with the name only the speaker knows), e.g. 'blurby'. The hearer hears the colour name, and tries to match it with a colour tile it can see. It then points to the colour tile it thinks the speaker meant, and the speaker points to the colour tile it actually meant. The hearer then adjusts its internal representations to fit what it has learned. This involves creating a new 'node' placed in a colour map at the location of the colour tile, labelled 'blurby'. From now on, the hearer can recognise any colour close to this as a 'blurby' colour. Over many exchanges like this, individuals come to share the same names for the same colour tiles.

However, no studies have looked at the additional problems produced by bilingualism. What happens when there are two languages being used? Not only can there be two labels for each type of colour tile, but the boundaries between lables can be different (for example, Welsh is one of the many languges that only has one word for the range of colours extending from green to blue).

In fact, no models even allow the possibility of having two labels for the same percept (labels are selected by calculating distances of label foci in a perceptual space). Additionally, most models implicitly assume a mutual exclusivity bias - if agents experience a new label, they assume it refers to a division of the perceptual space that they don't posses and adjust their perceptual space accordingly. However, agents could not necessarily assume this if they were learning two languages.

The solution is not as simple as creating two separate colour maps for each language, since all children have the potential to become bilingual there is no clear default. Also, being bilingual may affect the way you decide which object is being referred to. Studies into the Mutual Exclusivity bias have found that bilinguals are more likely to accept two labels for a single object, especially when they can tell the labels come from different languages (Au & Glusman, 1990; Healey & Scarabela, 2009; Byers-Heinlein & Werker, 2009).

Andrew Smith has investigated the problem of indeterminacy with computational models (Smith, 2005,2005b). The model is based on Steels' (1996) model of perceptually grounded meaning creation. Agents learn labels from each other which refer to perceptions of a simulated environment. Because agents do not have access to each other's internal representations, they need some way of overcoming indeterminacy when faced with a label they do not understand. Smith (2005) suggests that this can be achieved by relying on individual-level representation constraints (perceptual biases) or inferential tactics. One of these inferential tactics is Mutual Exclusivity. Smith models mutual exclusivity by having the agents exclude from consideration all objects for which it already understands a word when perceiving a new word.

The challenge, in light of more recent findings concerning bilinguals, is to define a model which favours mutual exclusivity within languages, but violates it between languages. It's not obvious how this could be achieved, nor how to test it. From previous studies, it's clear that stable 'bilingualism' is unlikely to emerge in such models, so perhaps two systems need to be developed in segregation and then integrated at a test stage.

First, however, there is the problem of detecting which language a word belongs to, or realising that more than one language is being spoken at all. Smith's agents perceive the world through a number of perceptual channels. Although some sort of entropy model could distinguish between two sets of words, it's unclear why this would be useful except for deciding which language was being spoken. In other words, it would only aid the fitting of the model without really explaining how it works.

One way of aiding the learning of labels in bilingual contexts would be to increase the number of perceptual channels to encode the communication context. For instance, an analogue perception of the signal to model language membership cues from prosody (although this still has the same problem as the entropy solution above). Alternatively, some channels could encode a perception of the speaker themselves. Each agent would have perceivable features, analogous to facial features perhaps, which could be used to distinguish which agent was speaking.

The channels could also be extended to include the encoding of a pragmatic cue, enabling a modeling of the experiments in Au & Glusman (1990) and Healey & Skarabela (2009).

Perhaps, however, these models already achieve such phenomena. The models begin with an inherently bilingual society - each agent begins with a minimal, random vocabulary and conceptual divisions. One question is what factors lead to a single, shared system. Is it an inevitable outcome in a cultural system with limits on resources and exposure to stimuli? Or does a shared system emerge because of biases implicit in the model which disfavour bilingualism? There appears to be a gap in the literature anyway, possibly even a PhD-shaped one.

Au TK, & Glusman M (1990). The principle of mutual exclusivity in word learning: to honor or not to honor? Child development, 61 (5), 1474-90 PMID: 2245739
Byers-Heinlein K, & Werker JF (2009). Monolingual, bilingual, trilingual: infants' language experience influences the development of a word-learning heuristic. Developmental science, 12 (5), 815-23 PMID: 19702772
DEBOER, B. (2000). Self-organization in vowel systems Journal of Phonetics, 28 (4), 441-465 DOI: 10.1006/jpho.2000.0125
Grassmann S, Stracke M, & Tomasello M (2009). Two-year-olds exclude novel objects as potential referents of novel words based on pragmatics. Cognition, 112 (3), 488-93 PMID: 19616205
Healey, E. and Scarabela, B. (2009). Are children willing to accept two labels for one object? Proceedings of the Child Language Seminar. University of Reading.
Smith, A. (2005). The Inferential Transmission of Language Adaptive Behavior, 13 (4), 311-324 DOI: 10.1177/105971230501300402
Smith, A.D.M. (2005). Mutual exclusivity: communicative success despite conceptual divergence Language Origins: Perspectives on Evolution

Sitting in the stand

2010-01-04T15:47:00.003Z

I noticed today a bilingual sign on Cardiff city FC's new stadium, reading "eisteddle Canton stand" (word order on welsh and English is usefully opposite so the sign can be read in welsh as "eisteddle canton" or in English as "canton stand"). What's interesting is that "eisteddle" means "sitting place", the opposite to "stand". As someone pointed out, though, "stand" comes from "grandstand", refering to the building rather than the action. Still, it's interesting that coexisting languages can evolve to label the same thing with labels that oppose each other in other contexts.

Ar y Tracs 2

2009-12-23T01:01:00.001Z

Tonight I watched the 'making-of' documentary for Ar y Tracs (see my previous blog). Co-Author Catrin Dafydd spoke about the importance of bridging the gap between Welsh and English in order to capture the complexities and humor that arises from bilingualism in Wales. Also interesting was to see her and Ruth Jones discussing their co-authorship. Here's a quote from Dafydd ('English' marked in bold for readablility):

"Ac on i'n dweud 'Oh what about this', ie, 'galla hi dweud hwn' you'd say 'wouldn't it be simpler if she just said this', a wnest ti cynnig pethau yng Ngymraeg. An' actually the perspective of someone who's learned the language really helped the script I think because hopefully it's going to be something, ti'n 'mod, really accessible i ddysgwyr ac i bobol sy ddim mor hyderus, confident, yng Nghymraeg."

(And I'd say 'Oh, what about this', ye, 'she could say this' you'd say 'wouldn't it be simpler if she just said this', and you'd suggest things in Welsh. And actually the perspective of someone who's learned the language really helped the script I think because hopefully it's going to be something, y'know, really accessible to learners and to people who are not so confident, confident, in Welsh)

A couple of interesting points here. First, a minor point about the difficulty of deciding which language is being spoken. In the final phrase (it's going to be something, ti'n 'mod, really accessible i ddysgwyr...), it's difficult to tell when Dafydd has swiched to Welsh. The first 'Welsh' words are 'i ddysgwyr', but there's evidence to suggest that switching has before this. First, 'really' is considered a loan word by many. In fact, an analysis of Dafydd's speech would be necessary to determine if she considered 'accessible' as a Welsh word, too. Furthermore, one would need to survey Dafydd's use of Welsh and English interjections to determine whether '"ti'n 'mod" (y'know) was a marked cue to language alternation.

The second point is about using other people's codes. Right at the end, Dafydd sub-titles her Welsh (hyderus, confident, also just before this quote with "... dwyiethyddiaeth, bilingualism ..."), presumably for the benefit of the less-fluent Jones. In fact, you can see Dafydd struggling with having to monitor her speech to make sure it's understandable while also trying to make a complicated point. Finally, she switches into English ('And actually the perspective ...'), which they both understand. Once Dafydd has collected her thoughts, she switches back into Welsh for the remainder of the sentence ('really accessible i ddysgwyr...').

My research focuses on why humans have a great capacity for bilingualism. The quote above demonstrates that it may be easier to adopt the 'code' of another than be constantly monitoring your own code to make sure they will understand you.

Bilingualism in Ar y Tracs

2009-12-20T22:47:00.010Z

I've just finished watching the new comedy from Ruth Jones, co-writer and star of Gavin and Stacey. Ar y Tracs (On the Tracks) is a Welsh-language production for S4C (Tidy Productions/ Green Bay) following the lives of a train crew over the festive period. Ruth Jones, who is learning Welsh, co-wrote the script with Catrin Dafydd who has written for the Welsh-language soap Pobol y Cwm.

Having been away from Wales for a while, it's always a bit weird to see people speaking Welsh on TV, but this programme was notable for its extensive use of English. Characters moved in and out of thier two languages all the time, much more than you'd usually see on Pobol y Cwm, or S4C's answer to Hollyoaks, Rownd + Rownd.

As well as switching to talk to non-Welsh speaking characters, and idiomatic borrowing ("Does dim second chance da fi nawr - I've had my lot") there was plenty of inter-sentential language alternation:

"Pan ti’n mynd trwyddo i Big Brother 10, bydd, like, masif support i ti ar y we yn barod."
(When you go through to Big Brother 10, there will be like massive support for you on the web already)

Also, a good use of marked language change to emphasise dramatic turns:

A: Pwy yw Billy Bricks?
(Who is Billy Bricks?)
B: Billy Bricks was my father.

There was also a nod to Gavin and Stacey with a typical Welshifying of the catchphrase "Beth sy'n occuro?" (What's occurring?). It was really nice to see naturalistic speech. Carolyn Hitt from the Western Mail voices the same opinion.

S4C 's language scheme states that "a substantial proportion of the programmes broadcast ... must be in Welsh, and, in particular, that those programmes which are broadcast on S4C during peak viewing hours are mainly in Welsh." However, nowhere in S4C's language policy (or any other official document I've seen) is 'Welsh' actually defined. If a programme had half English words and half Welsh, would this count as a Welsh-language programme? What would a word with a Welsh stem and an English affix be counted as? What if Welsh were always the matrix language? Will broadcasters have to start hiring linguists to check their statutory obligations? I hope so (my persistence with this blog shows how far my PhD has progressed).

A possible defence for S4C in light of its language duties is to recognise that the majority of 'Welsh speakers' in Wales use Welsh as part of a cohesive Welsh-English code. Since Ar y Tracs was certainly in peak time, this may be the first endorsement by a public institution of a code-based view of communication! There's hope yet.

Kovacs & Mehler Followup

2009-12-16T12:20:00.006Z

Last week I discussed Kovacs & Mehler (2009), here. I now understand that the participants in the study were bilingual in Italian and Slovenian. Here's an update on the data from the last post and also data for all onsets (not just 3 syllable words):

It looks like there is no reason to believe that Slovenian-speakers would have more experience of ABA structures than Italian speakers. For some reason, the proportions of ABB syllables seem to shrink in comparison to AAB and ABA when taking all onsets into account. However, the distributions over nouns, adjectives and verbs for English, Dutch and German are a lot more even when considering all onsets, suggesting there is not a huge difference in the way different languages indicate syntactic class using syllable structures.

Kovacs, A., & Mehler, J. (2009). Flexible Learning of Multiple Speech Structures in Bilingual Infants Science, 325 (5940), 611-612 DOI: 10.1126/science.1173947

A mixed week for Bilingualism in Canada

2009-12-14T16:18:00.003Z

This week has seen both the good and bad side of language policies being played out in Ottawa, Canada.

On the negative side, there has been some debate over bilingualism in official roles. There's been a call for a requirement that the Fire Chief of Ottawa be bilingual. Meanwhile, a monolingual post-office worker is fighting to keep her job because she is not bilingual, prompting locals to organise a petition (here).

On the other hand, there are also inclusive bilingual policies being implemented. Vancouver is hosting the winter Olympics next year, and this week it was announced that it will be a bilingual experience:

The Vancouver 2010 Olympic and Paralympic Winter Games provide an unprecedented opportunity to showcase our unique Canadian identity to the world. ... the Vancouver Organizing Committee has devoted a great deal of time and resources to ensure these Games reflect our country’s world-renowned diversity, including its linguistic duality.

Ghost in the Shell Graffiti

2009-12-13T15:15:00.000Z

I seem to have an obsession with Ghost in the Shell! Here's some stencils I made (see my first one here):

Here's a Tachikoma, an old stencil that I re-sprayed today!

Major Motoko Kusanagi. I messed up the face, so decided to do some kind of light-explosion thing. This is my first attempt at painting with acrylic. Mixing colours was fun, but I realise I didn't really understand layering, and I was still painting in a digital frame of mind.

A Laughing Man icon! The quote is from J.D. Salinger's Catcher in the Rye.

Here are some other stenicls I found on the internet. There were suprisingly few, though.

An amazing stencil from Cyberdelics.

By Actionconformist.

A cheeky stencil plan by Okto.

A much better Laughing Man icon from mmoroca.

This one kind of counts. From Pierre Huyghe, Philippe Parreno.

Terms of Endearment

2009-12-11T09:31:00.000Z

I've been thinking about nicknames and pet names. The people that mean the most to us usually have more than one name (see my post on Sliders). This appears to contradict the mutual exclusivity bias to have only one word for each object. That is, unless we see names as social tools we use to manage our relationships with people. Morgan et al. (1979) discuss the importance of managing social relationships with nicknames, for instance using a full name with your parents and a more informal diminutive with friends. Andersen (2006) discusses nicknames as adaptive innovations that serve the speaker’s emotive expressiveness. As Jhumpa Lahir puts it in The Namesake:

In Bengali the word for pet name is daknam, meaning, literally the name by which one is called by friends, family, and other intimates, at home and in other private, unguarded moments. Pet names are a... reminder that life is not always so serious, so formal, so complicated. They are a reminder, too, that one is not all things to all people... Every pet name is paired with a good name, a bhalonam, for identification in the outside world. Consequently, good names appear on envelopes, on diplomas, in telephone directories and in all other public places.

Perhaps, then, there is a way of linking bilingualism to social grooming theory. That is, language has taken over the social role of paying attention to significant others and the more 'effort' you put into innovation, the more attentive you are perceived to be.

Processing two structures simultaneously

2009-12-09T14:04:00.007Z

Although public opinion is still coming round to the idea, Bilingualism does not, in fact, impede cognitive and linguistic development, but may enhance it. Bilinguals have been shown to be more aware of the use of words in social contexts (Rosenblum and Pinker, 1983), they are better at taking other speaker’s perspectives (Genesee et al., 1975), and better at monitoring the knowledge state of others (Genesee et al., 1996). Recently, Kovacs and Mehler (2009) showed that bilinguals are more flexible at processing linguistic structures.

Kovacs and Mehler (2009) run an eye-tracking experiment on infants. Children had to learn to associate three syllable words with either AAB (e.g. 'babaka') or ABA ('bakaba') structures with a stimulus appearing on either the right or the left side of the screen. Bilinguals successfully learned to associate both structures with the correct side, while monolinguals performed equally well for AAB structures, but could not learn the associations for ABA structures. (See Ed Young's blog for another analysis). This is a bit confusing - it's not just that monolinguals were worse than bilinguals at processing ABA structures, monolinguals actually scored negatively!

Bilinguals have a wider range of input than monolinguals, and so may be more used to different linguistic structures. Bilinguals may be better at processing two linguistic structures simultaneously. Alternatively, monolinguals may have refined their processing to fit their input language.

Part of Kovacs and Mehler’s argument relies on AAB structures being easier to process, referring to Gervain et al. (2008). Presumably it’s true that this is because less memory is needed but Gervain et al. ﬁnd that ABB structures, not AAB, are easier to process than ABA structures in monolinguals. One question is whether this processing beneﬁt is language-speciﬁc. If it is, then the result may reﬂect bilingual’s experience with ABA strucutres. If it is not, then the result may show either a faster maturation of processing abilities in bilinguals, a slower maturation of processing abilities in monolinguals or a difference in investment of resources in the domain of word learning.

Let's look at some data! Proportions of various structures of 3 syllable words were gathered for English, Dutch, German (CELEX), Mandarin Chinese (CC-CEDICT, taking tone into consideration) and Italian (CoLFIS, although orthographic and automatically parsed for syllables):

The graph above shows that there is not much variation between proportions of word forms with AAB and ABA structures within a language. This suggests that bilinguals do indeed have a processing advantage. However, there is variation in the number of tokens of AAB and ABA across languages, and large variation in the proportions of word forms with ABB structures compared to other structures. If this is the case, then bilingual infants may be exposed to (and therefore be more effcient at processing) a greater range of structures, which may be an additional factor in breaking the Mutual Exclusivity bias.

Furthermore, the proportions of words which conform to either AAB, ABB or ABA structures are very small. Why, then, are these structures used frequently in infant-directed speech (Ferguson, 1983)? A possible answer is a kind of explicit demonstration of linguistic structure. At any rate, Kovacs & Mehler's paper is interesting, as is it's companion paper, described here.

Below, I've split the data from the graph above by counts for Adjectives, Nouns and Verbs. Interestingly, German, Dutch and English have different distributions. For instance, a new word with an ABA structure (e.g. 'bakaba') may be interpreted differently by different speakers (rationally, ignoring heuristics based on morphological cues). An English speaker would assume it was an adjective, a Dutch speaker would assume it was a verb and a German speaker would assume it was a noun:

This would show that different languages have different ways of cuing children into the meanings of their words. Unfortunately, perhaps because of the very low numbers of examples, when I tested some English and Dutch speaking friends on this, they basically chose randomly. Another big, unfounded linguistic theory.

(also, see my followup post here)

Kovacs, A., & Mehler, J. (2009). Flexible Learning of Multiple Speech Structures in Bilingual Infants Science, 325 (5940), 611-612 DOI: 10.1126/science.1173947

Bilingual Puns in Bali

2009-12-09T11:58:00.001Z

I came accross an old paper by Joel Sherzer on bilingual puns and word play in Bali. There are several languages in use in Bali, including Sanskrit, Old and Middle Javanese, Balinese (including the various levels - alus 'refined', biasa 'ordinary', kasar 'coarse' etc.), Indonesian and English. Most people speak many of these, and the interplay between them is a common feature of dialog. Here's some examples:

X (to Y, in Indonesian): Sudah siap? 'Are you ready?' (lit. 'already ready')
Y (in Indonesian): Sudah ayam 'Already a chicken.'

Here, siap, which means 'ready' in Indonesian, means 'chicken' in Balinese Alus. Also, ayam, means chicken in both Indonesian and Balinese Alus.

X calls out: Wayan mejalan cara taluh 'Wayan walks like an egg'.

Taluh is Balinese biasa for 'egg'. 'Egg' in Balinese alus is adeng. Adeng in Balinese biasa is 'slow'. That is, Wayan is walking slowly. People can also conduct entire conversations where the meaning is actually based on puns:

X: Mekunyit di alas? 'turmeric in the forest?'
Y: Ketemu '(type of) spice'

Here, X is asking Y if they have a girlfriend, since Ketemu is both a kind of spice and 'acquaintence'. This might just seem anoying, but it is by far the least complex punning interaction. Here's a section on popular ways of saying 'goodbye':

Here are some examples involving the sound similarity between Balinese siu 'one thousand' and Eng. See you. A person may say Siu surat, lit. '1000 letter', but a play on Eng. 'See you later', in which B, I surat 'letter' is a pun on Eng. later. Or a person may say Siu berjumpa, with Indonesian 'meet, see'. The use of meaning equivalences in different languages to go nowhere referentially is also the basis for such comebacks as Siu one thousand, based on the fact that siu is Balinese for 'one thousand'. Or a person may say Siu seribu, in which siu stands for Eng. 'See you' or Balinese siu 'one thousand', and seribu is Indonesian for 'one thousand'.

The most baroque and recherche in this group of mock leavetakings is Siu satak, lit. 'one thousand two hundred'. Again this takes off from the similarity of Balinese siu and Eng. See you, but added to this is the fact that '1200' can also be expressed as nem bangsit, lit. 'six two-hundred' - in which bangsit, with the m/b interchange seen above, sounds like mangsit 'to stink'. Once again the play is not on an uttered word, but on an imagined or presupposed word.

WHAT?
Aparrently, types of pig-latin are farily common including:
Inserting syllables with vowel echoing
Deleting all but the first Consonant-Vowel-Consonant sequence of each word
Reversing syllables
Reversing phonemes

WHAT WHAT WHAT?

Dubloons, loonies and moonies

2009-12-08T09:41:00.001Z

Recently, I've been thinking about the Mutual Exclusivity bias - the tendency children and adults have to assume one meaning has only one associated word (in other words, to avoid homonomy). This bias seems to work against bilinguals, who have two words for many things. I'll get round to discussing this at some other point. In the meantime, here's an extract from an interesting article by Neil Wick on the tendency for new words for new meanings to converge on a single, conventional form. Wick charts the course of nicknames for the newly introduced $2 coin in Canada. It's worth noting that money is not a normal object - its meaning remains constant in all contexts. However, it's quite a good story:

"Speculation on a name began as soon as the new Canadian coin was announced in the February, 1995, federal budget. On March 3, The Globe and Mail reported that “the ink was barely dry on Paul Martin’s new budget before lock-up wags were bantering around a new name for the $2 coin” and doubloon was mentioned among other suggestions (“We like Ike” 1995). Both The Globe and The Star printed numerous letters hotly debating myriad possible names. One thing seemed beyond debate—writers felt strongly that a name would emerge. The Toronto Star confidently predicted on October 7, 1995, “It will acquire its own nickname” (Aaron, 1995). The Word Play columnist at The Globe implicitly acknowledged a lexical gap on February 24, 1996: “Writers of letters to The Globe and Mail have been diligent about filling this void” (Clements, 1996). One reporter even suggested that the coin needed a name. On February 16, 1996, a Star business reporter stated “All it needs is a nickname— and there is no shortage of suggestions” (Hemeon, 1996). The Mint would have no part in settling the debate. A Globe story on March 23, quoted a Mint spokesperson: “We’re not in the habit of giving names to any of our coins. For us a 10-cent piece is a 10-cent piece.” Characterizing such coin names as too unprofessional to be used by Mint workers, he declared, “The public will have to sort out [what to call the coin] on its own” (Grange, 1996). In spite of the heated debates which still continued after the coin’s launch, a consensus was already forming a month before the launch, as evidenced by responses to The Star’s request for readers to phone in their name suggestions (Stefaniuk, 1996a). Among 57 names and variations submitted by readers, four stood out. Teddy had 11 votes, Toonie/Twoonie/Twooney had 10, and two variants were tied for third place with 9 votes each: Doubloon/Doubloonie and Moonie (Stefaniuk, 1996b). Aaron (1996b) lamented on March 9, that “the horrible term ‘twoonie’ seems to have an edge in public acceptance.” On February 19, 1996, the official launch date for the new coin, Freeman (1996) wrote in The Globe that the new coin “has already picked up a string of unofficial names such as toonie, doubloon, bearbuck, blooney, Doosie and Loonie II.” Meanwhile, a March 14 Star article about panhandlers’ experiences with the new coin (DeMara, 1996) used the word toonie 10 times without remark, prompting an angry letter accusing The Star of “trying to shove the word ‘toonie’ down our throats” (Moshinsky, 1996). “Over here, it’s poly (polar bear – see?) or polies; always was, always will be, The Star’s decree notwithstanding,” the reader wrote. This reader may have overestimated the influence of the paper on public consensus. In fact, if the name depended on a Star decree, dubloon or dubloonie probably would have won out. This was the name used most by The Star in early stories, it was preferred by the Coins columnist (Aaron, 1995) who found toonie to be a “horrible term” (1996a) as already noted, and it was even the personal choice of the chief lexicographer of the new Gage Canadian dictionary (Grange, 1996). Toronto Star art critic Christopher Hume on March 21, 1996, recognized that the lack of a stable name put the two-dollar coin in a different category than that of the one-dollar coin introduced nine years earlier: “By contrast, the loonie has become part of the culture. The word has entered the vocabulary” (Hume, 1996). He called the two-dollar coin “still, annoyingly, nameless” yet he matter-of-factly called it a toonie twice in that same column. By the end of August, Aaron was describing the coin as “affectionately called the ‘toonie’” (Aaron, 1996a), and by September, The Star acknowledged that “its colloquial name ‘toonie’ is part of the vernacular” (Vincent, 1996). In March of the following year, Kesterton (1997) wrote in The Globe that “Fairly quickly and dismissively, Canadians have come to call the $2 coin the ‘toonie,’ despite the many clever terms that were suggested by word mavens …”. Clearly, cleverness alone was not enough. The winning candidate was efficiently short—less than three syllables as with the other coins’ names, incorporated an allusion to the word two, rhymed with loonie, and recalled the familiar collocation looney tunes. Importantly, it started to build momentum in public acceptance early in the process and once established in a few speakers’ lexicons, there was little chance that those speakers would accept alternatives barring major pressure from another stronger group of speakers."

Biculturalism

2009-12-02T13:00:00.000Z

I came across a series of films on You Tube with famous faces talking about growing up with two languages, and two identities, from the Equality and Human Rights Commission. For some reason, they're all subtitled in Welsh, even though the speech is in English. The famous faces include Sanjeev Bhaskar, Gok Wan and Boris Johnstone.

Hawkers 2

2009-12-01T10:46:00.000Z

In October I wrote about a Big-Issue hawker selling his wares. A recent article from Edinburgh's student newspaper, The Journal, interviews him, here. In addition to his 'Can I interest you in the Big Issue?', he has recently added 'Don't be shy come and buy/try!'.

Where the Welsh Things Are

2009-11-30T15:27:00.000Z

Maurice Sendak's classic children's story 'Where the Wild Things Are' has been made into a movie and is released in the UK this week. I, like millions of other children, read this as a child. However, I will feel an extra level of betrayal to those who don't agree with the movie changes to the voices/plot/characters, because I read it initially in Welsh.

The welsh speaking characters in 'Gwlad y pethau Gwyllt' always felt closer, more special to me, in a way the characters in English books did not. Indeed, they may have been more dear to me literally - I discovered that my Welsh version of 'Where the Wild things Are' is a collectable now, selling for $450 dollars on Amazon! I should really try to find it...

It's not the first time this has happened. I remember seeing Superted in English for the first time (Superted was originally in Welsh), and feeling like he had abandoned Wales and the fight to 'achyb yr iaith' (rescue the language). I also remember being surprised that my English speaking friends also knew about Fireman Sam.

I wonder if other children around the world have been similarly confused. In the end, I guess I shouldn't have been that surprised that my childhood heroes were bilingual, but why did they never code-switch?

Talking of code-switching cartoons, I can't help including possibly the most complicated example of code-switching ever, from Disney's The Prince Of Egypt:

Conceptions of bilingualism in Canada

2009-11-27T16:51:00.001Z

I just finished watching the documentary 'Incident at Restigouche' about tensions between the Canadian native Micmac Indians and the Quebec government over fishing rights in 1981. The Quebec authorities raided the local reserve because, in their eyes, they were over-fishing the river. The Micmac deny this, pointing out that sport fishing alone took a greater number of fish each year. I was linked by The AQ's blog on the Micmac language and identity, here.

Several things made me upset. First was the labelling of the Micmac as 'not bilingual' because of speaking English and Micmac, not English and French.

Secondly, I was particularly struck by what Quebec Minister of Fisheries, Lucien Lessard, was heard to say:

"You cannot ask for sovereignty, because to have sovereignty, one must have one's own culture, language and land."

First of all, the Micmac have all of these, so it's not clear what on earth Lessard was thinking. It's clear, though, that people's conceptions of other languages and how they relate to culture and sovereignty can be radically different.

Bayesian Bilingualism

2009-11-24T10:53:00.001Z

I've been wondering about Bayesian models of language learning and bilingualism. Models such as Griffiths & Kalish (2005) assume learners have probabilities for hypotheses of the structure of a language in a large hypothesis space, based on utterances heard. The posterior probability represents the learner’s model of a speaker’s language (compatible with a view of trying to learn the parents’ Medium). Two methods drive convergence to a best hypothesis in the learner: The MAP (maximum a posteriori) process assumes the maximally probable ‘language’ and only produces strings created by that ‘language’. The sampling approach (SAM) does not rule out any nonzero probability hypothesis and may produce mixed strings occasionally.

In a monolingual environment, MAP should be most efficient, but SAM is better for acquiring more than one language. A sampling approach also models observations of better task switching but worse inhibition in bilinguals than monolinguals. This may be another factor in the differences between monolingual and bilingual development.

However, I'm not completely sure about the maths, and suspect that MAP can define a best hypothesis over any number of 'languages', so they may be equivalent.

Bilingualism in Singapore

2009-11-18T00:14:00.000Z

Singapore is certainly multilingual. It has four official languages, and Ethnologue catalogues 21 different languages and dialects, all within about 5 million people.

The latest post is on Language Log is on Bilingualism in Singapore, charting the dubious theorising of Minister Lee Kuan Yew (for more on Lee, see LeeWatch.info). Lee essentially forged an education plan based on the idea that people only have so much 'space' to store languages, and so bilingualism can only be bad. Surprisingly, this was not so far from well received Linguistic theory until Martin-Jones and Romanie (1986) argued it was 'half-baked'.

However, the minister seems to have recently changed his mind about bilingual education for the better. Maybe it'll trickle down to the guy I met the other day who congratulated me on my very good English despite my Welsh medium education.

Laughter

2009-11-17T18:09:00.001Z

Last week I heard someone laugh, and I thought it was such a good laugh that I would use it from then on. It was sort of halfway between a cackle and a guffaw - definitely a mocking, cruel, delighted burst of air.

This reminded me of another concious adoption of a cultural trait that I read about recently from Papua New Guinea: McElhanon describes a community meeting in which one group decided to change their word for ‘no’ in order to distinguish themselves from another group (Kulick, 1992). Although there are many examples of people changing to conform, it's not often you find such an organised move away from the norm. I couldn't find out how successful the change was, though.

Surprisingly, my own adoption seems to have worked, and I now involuntarily use my new laugh quite a lot. Laughter, it seems, is infectious.

Check, Fold, Race

2009-11-15T14:04:00.000Z

Edinburgh Skyline

2009-11-14T17:26:00.001Z

Me and Keelin just made a stencil of the Edinburgh skyline from Princes Street. I was really pleased with it, and will definitely use sponging again. The whole thing only took about 4 hours.

Here's the original panorama. I cut out the boring trees in the middle.

The contrast on the castle wasn't great, so I ended up splicing in a separate image:
Desaturate, boost contrast, find edges, print:

Cut out!

Stencil Ready:

Sponge acrylic paint onto wall:
Stencil done!

Ha Long Time Coming

2009-11-13T15:04:00.000Z

David Graddol predicted in 1996 (theory quantified by Lupyan and Dale, 2009) that, as more people learned English as a second language, native English speakers would lose their grip on the language. Indeed, there are probably far more second language speakers of English now than native speakers, so the non-natives have the power to change the language to suit themselves.

This was highlighted on a friend's travel blog recently (I've been enjoying living vicariously, especially when there are puns involved in the titles). The latest post in Eric's South East Asia blog finds the protagonist on board a ship, floating between the myriad of islands in Ha Long Bay, Vietnam. Incidentally, I have to agree that it is one of the most beautiful areas I have ever seen. Having signed up for an English tour, Eric is surprised to find that he is the only native speaker of English. However, he found himself translating between 'versions' of English. There seemed to be non-reciprocal intelligibility between them: Eric could understand them all, but they had difficulty translating between themselves. Indeed, one traveller resorts to using his iPhone for translation (see my post on Lingua Tecnologia).

So it seems that, instead of a sprawling continent, English may be eroded by the seas of time into thousands of tiny little islands.

Ghost in the Shell

2009-11-12T18:10:00.001Z

This was my favourite stencil. Unfortunately, the original got lost in a move and I haven't had the heart to do another. It's inspired by Shirow Masamune's original Ghost in the Shell comic. In this scene, the hero glimpses the birth of a new, intelligent life form. The thing in the middle is some kind of neural network, and descending onto it is an angel (the shape pointing down is the shade on a foot, some people don't notice it at first, but after it's pointed out, you can't help seeing it).

The idea is similar to the Sistine Chapel ceiling fresco, where God and Adam reach out towards each other. Michaelangelo captures the question of man's relationship with God. Masamune questions the relationship between man's body and man's spirit.

The medium here also hints at a direction for answers: Stencils are made up of use small, abstract, isolated shapes. However, in a particular configuration, they create an impression of a unified image to human perceivers. However, not all shapes are allowable - you can't have a stencil with free standing opaque parts (e.g. the outline of a full circle). In fact, even convex shapes decrease the integrity of the stencil. So, even as the whole influences how one interprets the parts, the parts influence what that whole can be.

Codeswitching as a Move to Markedness

2009-11-11T15:45:00.001Z

One advantage of having two languages is having an extra tool with which to avoid ambiguity. For example, in English, ‘Thirteen’ and ‘Thirty’ are often confused, while in German ‘dreizehn’ and ‘dreissig’ are more different, while in Chinese ‘三十’ and ‘十三’ are very different. Montanari (2008, pp. 622) gives an example of this tactic in a trilingual child (KAT) interacting with their grandmother (GRA) in Spanish and Tagalog:

%sit : KAT and GRA are engaged in book reading
*KAT : [‘ota].
%gls : pelota
%eng : ball in Spanish
*GRA : ¿botas ? zapatos ? zapatos.
%eng : boots ? shoes ? shoes.
*KAT : bola bola !
%eng : ball in Tagalog
*GRA : ah la pelota ahí detrá s, ahí está la pelota.
%eng : ah the ball right behind, there is the ball.

Because the child cannot pronounce the ‘pel’ of ‘peloa’ (ball), their attempt is confused with ‘botas’ (shoes). Instead of attempting the word again, or using pragmatics, the child uses the word in a different language. This makes it easier to pronounce and thus easier to understand. Perhaps, then, some codeswitching can be accounted for by this tactic.

One might assume that the optimal strategy, given two different languages, is to switch at every word. However, individual languages tend to display a move to markedness (Shillcock, Hick, Cairns, Chater & Levy, 1995). This principle is ‘that when consonant interactions introduce phonological ambiguity, the ambiguity introduced is always in the direction of a less frequent phoneme’ (Tamariz & Shillcock, 2001). That is, frequently occurring words should be optimised for pronunciation within a language, while words from another language will be free from this pressure. This suggests that frequent constructions (e.g. Noun Phrases) should be most salient in the same language. However, at larger phrase/constituent boundaries, where the probability of words co-occurring is less, words from other languages may be more salient. Code-switching phenomena such as Myer-Scotton’s embedded language frames may fall out of this interaction.

A modelling approach could be used to investigate this. A list of cognates and sentence templates in two languages will be required. Sentence templates will be filled with words from either language, based on maximising the phonetic distinctness of the sentence. This will be calculated using Markov Chain assumptions, with words as nodes and transition costs as the phonetic difference between the last phone of the current word and the first phone of the next word. To model this for children, extra costs could be imposed on transitions to words with complex consonant clusters.

This will produce sentences which are maximally phonetically distinct. Inferences about the choice of language could be drawn over many sentences and many sentence types, with particular attention being paid to constituent boundaries.

Mixing into weaker language

2009-11-10T12:40:00.000Z

Studies above show that bilingual children can differentiate between their languages and show sensitivity to their interlocutor’s linguistic abilities from a very early age. Yet it is still implied that mixing occurs for qualitatively different reasons to adults. For instance, Cantone & Müller (2007) suggest that children mix more often into their weaker language. However, adults also have lexical gaps in weaker languages. The assumption of separate lexicons is weakened in a draft of Cantone & Müller’s paper:

“… one word und two word utterances …” (Cantone & Müller, 2007b, p.8, my emphasis).

Cantone & Müller’s theory could explain this mixing – that is, they are mixing into the language they are less ‘ready’ to speak (both authors are stronger speakers of German). The example is harsh, but demonstrates that it is unclear why their theory applies to developing children alone. The observation that both children and adults mix more into a weaker language is not surprising, and may only be a quantitative difference.

Modeling Bilingualism

2009-11-09T13:43:00.002Z

When children are brought up speaking two languages, they often go through a stage of 'mixing' where they appear to be unable to separate their languages. For instance, a Welsh word might be inserted into an English sentence: As an example, when I first realised the implications of death, my parents told me that I cried and said "I don't want to go into the pridd" (earth, dirt).

Several theories have been put forward to explain this. Firstly, I may simply not have known the word for 'dirt', and had to rely on a word in another language. Back then, Welsh was probably my stronger language, so this would be an example of mixing into my weaker language. Alternatively, I had not yet learned to tell the difference properly between Welsh and English.

However, both my parents speak Welsh and both languages are used, probably with quite a lot of mixing. Therefore, I may have known the English word, and been aware that I was mixing, but I knew that using a bilingual code was permissible, given my interlocutors.

Indeed, Montanari (2008) finds that the child she studies mixes some words even when they know the word in the language of context. Does this suggest, then, that the child simply didn't know which words belonged to which language? I argue that this isn't necessarily the case.

Adults mix their languages for many reasons. In fact, it's often difficult to decide which language a word belongs to without a lot of context (e.g. 'zeitgeist'). Let's forget about languages for a minute and ask 'to what extent has the child acquired the communicative code of its parents'? By this, I mean how closely does the child's output mirror the parent's input?

To do this, let's look at Quay's (2008) study of a trilingual child. Japanese is the language of the environment, the father is strongest in English and also speaks Japanese and the mother strongest in Chinese and also speaks English and Japanese. Weekly recordings were made from 1;10 to 2;4 years. The utterances of both the child and the parents were coded along with the addressee. The summary of the data is very detailed - containing the proportions of mixing between any two people in Japanese/English, Japanese/Chinese, Chinese/English and Japanese/Chinese/English.

Let's model the child's mixing proportions as a function of the parent's mixing proportions. Each cell in the table below contains the correlation between the model’s predictions and the child’s actual mixing proportions. The first two models use the mother and father’s data separately. The third model is an additive model which combines the parents’ utterances and the fourth uses the difference between the parents’ mixed utterance types. The difference model was provided as a conceivable, but unlikely model. The correlations in the first column correspond to a model using the total input, whereas the last two columns correspond to a model using only utterances directed to the child (direct) and utterances directed to the other parent (indirect).

Although the mother spends more time with the child than the father, the total mixing behaviour of the child is equally predicted by the mother and the father. However, the best model is an additive model of the direct utterances to the child. That is, the child's output is closest to a model which tries to imitate the mixing behaviour of both parents.

Interestingly, the highest correlation between the mixing proportions is between the parents (0.999), which is nearly perfect. Perhaps, then, the child is simply trying to acquire the adult’s mixing strategies or 'Code'.

We can look at the data in more detail by calculating the correlations between mixing proportions for each interlocutor separately:

When addressing the mother, the child's mixing proportions reflect the mother’s total mixing proportions better than the father’s and vice versa, indicating pragmatic differentiation to each parents’ mixing. When addressing the father, the child’s mixing proportions reflect the mother’s indirect input. This could indicate that the child is mimicking the mother’s interaction with the father. The opposite isn't true, but any mimicry may be masked since the child spends so much time alone with the mother.

These two analyses conclude that the child’s mixing reflects the mixing of the parents from a very young age. Modelling allows us to gain extra insights on the potential learning mechanism for the child, but it relies on detailed data, as in Quay (2008). The model could be taken further to include considerations of location, the societal status of each language and the parent's tactics (Negative evidence, implicit allowance of mixing, teaching of translation equivalents etc.).

Now for the ambitious, unfounded part: Considering a communicative code, there may be no qualatative difference between mono- and bi-lingual language acquisition. How, then, do bilinguals select words? One possible solution is to use a sort of Bayesian probability distribution over the linguistic, social and pragmatic contexts for each word that represents the best estimation of when to use a word. If a mapping between words and pragmatic and social contexts is acquired, a discrete mapping between words and ‘languages’ becomes irrelevant. This approach works equally well for acquiring one ‘language’, or several levels of tone or dialect.

In this sense, the ‘remarkable’ ability to keep languages separate (Costa & Santesteban, 2004) seems less remarkable and less specific to bilinguals: We don’t find it remarkable that an adult refrains from using terms of endearment during a boardroom speech.

This approach would be extended to syntactic acquisition by assuming that, as the mapping between words and meanings developed, strings of words themselves became a context which was encodable in the probability distributions of words. This is essentially a constructivist approach to bilingual acquisition: Before linguistic acquisition, infants first learn an embodied perceptual ‘language’ – an iconic mapping between form and meaning – which allows them to relate structure in the world to an interaction between sensory and motor activity. The mapping between structure in the world and symbolic, linguistic representations would build itself on top of this system in the same way as syntactic (Bernardini & Schlyter) and lexical (Nicoladis & Secco) acquisition can build on pre-existing structures.

Following from this, the ‘difficult’ bit of language acquisition is not the segmentation of strings into words or words into lexicons, but the initial segmentation of the world into functional concepts. The development of this more fundamental understanding of the world may be an additional factor in the qualitative differences between mixing in children and adults.

Languages and Poetry

2009-11-08T16:52:00.001Z

Languages

by Carl Sandburg

There are no handles upon a language

Whereby men take hold of it

And mark it with signs for its remembrance.

It is a river, this language,

Once in a thousand years

Breaking a new course

Changing its way to the ocean.

It is mountain effluvia

Moving to valleys

And from nation to nation

Crossing borders and mixing.

Languages die like rivers.

Words wrapped round your tongue today

And broken to shape of thought

Between your teeth and lips speaking

Now and today

Shall be faded hieroglyphics

Ten thousand years from now.

Sing—and singing—remember

Your song dies and changes

And is not here to-morrow

Any more than the wind

Blowing ten thousand years ago.

Good point, Carl. However, poetry may be a particularly bad way to make points about language change, as Paul Valery says in The Art of Poetry, "poetry can be recognised by its ability to get us to reproduce it in its own form: it stimulates us to reconstruct it identically."

On the other hand, although poetry has a small transmission error in terms of phonetic reproduction, the fidelity of conceptual interpretation may be a different story. Show me a class of high school English Literature students, and I'll show you eleven different, badly written interpretations.

Lazy Linking Friday

2009-11-06T16:16:00.000Z

Time for some lazy linking!

I've been listening to Archers of Loaf recently - especially the incredible ending to their penultimate album, White Trash Heroes. This is likely to be in my collection forever.

I have been utterly captivated by guitarist hironou2525: Videos of beautiful guitar playing together with links to mp3s and very high quality tabs. Obviously someone who knows a thing or two about cultural transmission. They have a blog here, although I couldn't read it. I especially liked I do:

The Speculative Grammarian is a collection of linguistic satire. There is a load of stuff there, but this set of puns caught my eye.

This recent paper by Novembre et al. is interesting - Genetic distance between people is correlated with the geographic distance between them. In fact, a PCA graph draws a pretty good map of Europe. I was especially interested to see the analysis of Swiss genes - they divide on primary language!

While Linguistic Exogamy (marrying someone outside your linguistic group) is common in liguistically diverse areas of the world (Papua New Guinea, Amazon basin) this analysis may suggest it wasn't practised so much in Europe.

Levels of Bilingualism

2009-11-05T12:50:00.001Z

How many people in the world speak more than one language? Probably the vast majority. In the USA it's estimated at 18% (US Census Bureau), in Canada it's about 34% (Statistics Canada) and in the EU it's about 66% (European Commission). But getting data is hard - even in countries with the infrastructure to support a large scale census, the issue of bilingualism is often not prioritised. The metric of number of languages spoken in each country (linguistic density) has been used (e.g. Nettle, 1999), as well as the number of neighbour groups (Lupyan & Dale, 2009) and is probably correlated, but is not the same as bilingualism.

So, maybe we can estimate a different way. The Ethnologue has data on the estimated number of speakers for each language within a country, along with the number of people in a country. Subtracting the number of speakers from the number of people gives, in theory, the maximum number of bilinguals in a country.

Maximum Number of Bilinguals =
total number of speakers for all languages – total number of people

For example, if a country has 1 million people, and 500,000 speakers of language A and 750,000 speakers of language B, then 250,000 must be bilingual (if there are no other languages spoken). The figure below shows the ratio of speakers to people with darker areas indicating higher levels of bilingualism (data from Ethnologue, created with R):

click for larger image

As expected, the data is not good enough to warrant a proper analysis. The number of speakers is underestimated (total population of world = 6 billion, total number of speakers = 5.7 billion). 12% of entries in the ethnologue have no population data and for more than half of the countries the number of speakers is less than the number of people. One exception was Saudi Arabia, with a ratio of 9.4, possibly because 23% of the population are foreign nationals or, more intriguingly, because the majority of the population were nomadic until the 1960s.

At any rate, there appears to be no correlation with latitude (r= -0.1, t = -1.4, df = 197, p-value = 0.15) or longitude (r = -0.01, t = -0.28, df = 198, p-value = 0.8).

Ah well, back to counting people instead of numbers.
Gary Lupyan, Rick Dale (0). Linguistic Structure is Partly Determined by Social
Structure in Press

Genes vs Language

2009-11-04T16:38:00.003Z

A point about the difference between genetic and linguistic inheritance:

OneKind

2009-11-03T23:25:00.002Z

I can count the number of ad banners I have clicked on one mouse. But I had to click on this one:

[picture of cat and human] - "We're not that different, if you believe we're all OneKind".

As a linguist trying to understand the genetic basis of language in humans, I immediately thought "Uh-Oh".

The link took me to a sign up sheet with a picture of a dog and Paul O'Grady. This is what it said:

"Feelings. We all have them and so do animals. 9 out of 10 people agree. We’re all OneKind."

Amazing! A whole new way to approach the study of cognition in animals - tap into the 'Wisdom of Crowds' collective subconscious of the masses. It continued:

"Can an animal feel lonely, can an animal feel scared, can an animal feel pain? Common sense and experience have long implied that animals are capable of feeling. New research reveals that 91% of people believe this."

Who could argue with those statistics? I've always said that science gets things wrong sometimes - look at the classification of the tomato as a fruit! Surely this is having an adverse affect? Yes, OneKind tells me - Seals are having a bad time in Scotland, there are snares in the world and scientists are running experiments with monkeys.

What are OneKind doing about this? The only answer on the whole site? A petition! Which doesn't seem to be being sent anywhere!

"I believe that animals can feel.
I believe we’re all OneKind.
HumanKind.
AnimalKind.
OneKind."

I emphatically hope that my intrigued ad-click did not boost the ratings of a brainwashing cult. Not even brainwashing - just a group of people saying a meaningless statement, a sentence with a made up word, then three categories. And probably getting money and sympathy for it.

As an evolutionary linguist, I am very happy with the idea that we share many cognitive abilities with animals, and I'm not sure whether we'll ever find out whether animals have what people would describe as feelings. And, of course, I believe that all organisms are probably related at some level. What really annoys me, as a linguist, is that this site is exploiting a strange hypocrisy that, as Nettle & Romanie (2000) point out, people care about some obscure endangered species without caring that whole human languages are going extinct at a comparative rate. I'll be keeping my ad clicking in check from now on.

Cross-Dimensional Linguistics

2009-11-03T17:20:00.000Z

It's time for some Socio Linguistics! How many names do you have? What do you call other people? Would you use first names with some but never others?

So far, I have been Sean, Seanny, D, Lep and Monyn, but not yet Mr. Roberts.

Can we tell anything about people by the names they use? Let's look at a corpus. I have selected that pillar of linguistic research and popular 90s sci-fi tv program Sliders. Let's see what the four main characters call themselves while dimension hopping around alternative versions of San Fransisco and Los Angeles (from EarthPrime, with Regular Expressions):

In the graph above, Quin, Wade, Rembrandt and Professor Arturo are represented by dashed circles with the different names people use in squares. Arrows between a person and a name indicate that person using that name, with the relative frequency indicated by the thickness of the line. All arrows from the same origin sum to 100%.

So what can we learn? The professor and Wade have the most reciprocal relationship - they use each other's first name and title in similar proportions. This indicates a curtious respect.

The Professor and Quinn, however, have opposite proportions of first names and titles - this indicates that Quinn recognises his intellectual superior (although Quinn is much better, obviously).

Next, Quinn only calls Wade "Wade", while Wade also uses Quinn's first name the most, but also plays with "Q-Ball" (Rembrandt's favourite nickname for Quinn) and "Mr. Mallory". This pattern is typical of people who fancy each other, but, in the end, will never do anything about it.

The group's most complex relationship is with Rembrandt "Crying Man" Brown. Everyones uses his first name and affectionate diminutive "Remmy", while only the youngsters use his stage name and only the Professor uses "Mr. Brown". All this confusion is, lamentably, because even dimension hoppers can struggle socially around black people.

K**g Fu

2009-11-02T09:30:00.000Z

Etymologies

2009-11-01T18:25:00.000Z

Where do our words come from? English is Germanic, right? Ok, but what about words like Cappuccino, Revolution and Smorgasbord. Well, those were just 'borrowed' - they don't really count, since we're intending to give them back. But how valid is this view? Over the centuries, speakers have adopted words from all over the place, yet the diversity of the sources of words is under-appreciated.

Sounds like a job for ... etymology!

The general view of languages is that they are related like a family tree. English is seen as a Germanic language, along with Dutch and Flemish, while Welsh is seen as a Celtic language along with Irish and Cornish. The tree diagram below shows this idea, and gives the impression that the last 'common ancestor' of English and Welsh was way-back Proto-European:

However, this masks the complexity of languages and language change. A strict family tree marginalises the borrowing of words from other languages. For example, there are a huge number of 'English' words with roots in French, Italian and Spanish.

Hurford & Dediu (2009) encourage us to see languages as made up of sets of linguistic units (e.g. a word), each of which can have a separate ancestry. I wondered what this would look like, so I used the Online Etymology Dictionary to create one.

The Etymology Dictionary lists the heritage of English words, for example:

Cabin: 1549, from M.Fr. cabinet "small room," dim. of O.Fr. cabane "cabin" (see cabin); perhaps infl. by It. gabbinetto, dim. of gabbia, from L. cavea "stall, stoop, cage." Sense of "private room where advisors meet" (1607) led to modern political meaning (1644).

That is, the ancestry of 'Cabin' can be traced back through Middle French, Old French, Italian and Latin. Similarly, the word 'Tower' also comes from Latin, but via Old English. Crawling the website, the relationships for about 5000 words were processed. I used hypergraph to display them in an interactive hyperbolic graph. You can play about with it below, or visit here. Click and drag portions of the graph on the edges closer to the middle to explore. For some reason, it starts off zoomed in on Latin, but there's a lot of detail to the right (see here for abbreviations).

For ease of presentation, the graph is simplifed, with lineages of words between 'languages' first going through a language node. Also, Modern English words are not represented, but all contained within the 'Mod.Eng.' node.

Some bits of the graph are tree-like: Words with roots in Middle High German are only borrowed through (New High) German. However, in general, the graph is not tree-like at all. The lineages of English words have all sorts of routes through earlier languages. For example, words can come from Greek via German or French. And this is only for English words. Imagine etymological data from German and French was added.

Ok, so the graph is pretty useless for research - it's just way too complicated (part of the problem is that hypergraph is designed for trees). What I'm aiming at is questioning the idea of a 'language' as a stable set cut off from other 'languages'. We don't inherit a 'dictionary' from just two individuals, like our genes; we pick up individual words from a wide range of sources, and keep adding, borrowing and changing them throughout our lifetime.

Happy Scary Season

2009-10-31T13:48:00.000Z

All Hallows Eve, Hallowe'en, Halloween, Calan Gaeaf, Reformation Day, call it what you will, there are pumpkins being carved around the globe. Here's my attempt:

If you're wondering, the inspiration is from the film poster for Kurosawa's Throne of Blood.

Coke Maze

2009-10-30T17:11:00.003Z

I had an idea for a game: You build a maze out of cocaine, then have to snort your way to the end without dying. For an extra element of danger, some lines can be salt/anthrax.

The anti-coke campaign below had a similar idea. However, the maze seems unsolvable, so presumably the message is "Cocaine addiction is a problem, but there's no hope".

Entrance Music

2009-10-29T15:47:00.000Z

Many people have told me they know which song they want played at their funeral, but few have ever had a preference for songs to play at their birth. Ok, both are a bit strange, one you can't do because it's in the past and one you won't be around to enjoy anyway.

Of course, some would argue that the choice of music should really be left to the mother.
Dr. Jeni Worden has some advice on up-beat tracks for mothers, but can have some have mixed results, for example:

I spent the early part of my labour making a mixed tape of Motown, hip hop and rock to keep me energised. While this worked when I had been labouring relatively pain-free for eight hours, I wanted to destroy it as violently as possible when I was exhausted ten hours later.

This illustrates a difference between birth and death songs. While a 12 minute guitar solo might seem a bit indulgent at a funeral, there's a lot more scope for a birth song. Wagner's Ring Cycle wouldn't be totally out of the question. At any rate, choosing a song for the 'big moment' is still tricky. A balance must be struck between the rock and roll celebration of life and the choir-of-angels pretension.

In the end (or is it the beginning?), I'd like to come out rocking. Here are some songs I'd like to emerge to (unfortunately I've taken this rather seriously and none of them are funny puns):

1. It's all too much - The Beatles
2. untitled track (Kick out the Jams) - Rage Against the Machine
3. Le Cinquième As - MC Solar
4. The Blue Mask - Lou Reed
5. Cochise - Audioslave
6. Run Rabbit Junk - Yoko Kanno/Tim Jensen

Lingpisstics

2009-10-28T18:54:00.000Z

Sound symbolism describes a ‘natural’ connection between sound and meaning. For example, it has been noted that many words which sound the same also have similar meanings (e.g. Point, poke, pike, peg, pierce, prick, prod, probe). The idea is that some meanings (like the pointyness of the example words) lend themselves more naturally to certain sounds (like the explosive 'p' of the example words). It's been a contravertial theory, especially when applied accross languages towards explaining the origins of language (e.g. Rammachandran and Hubbard's synaesthetic bootstrapping hypothesis). More recently, the topic of 'ideophones' has take over.

It's fun to think of more examples, but here's a study I did a little while ago:

Sound symbolism may be felt most with very visceral words. Hurford (2007) suggests that words for ‘urine’ and ‘faeces’ should sound different across languages. Specifically, 'urine' words should have higher vowels to match the high pitch of pissing, (e.g. 'pee') while 'faeces' words should have lower vowels to match the bass sounds of the behind (e.g. 'poo').

Who said Linguists were always high-minded?

I tested this theory by finding Words for ‘urine’ and 'faeces' in 29 languages, including ones from Europe, Africa, North America, South America, Asia, Austral Asia, Australia and the Pacific islands. There was a bias towards selecting 'childish' or child-directed words over clinical terms (e.g. 'poo' over 'faeces') following arguments for the primacy of childish language (Cytowic 1995). Here are the results:

Words With...	Urine	Faeces
High Front Vowels Only	26	5
No high front vowels	7	32
Mixed vowel types	12	5
Total	45	42

There were significantly more words for urine with high front vowels than for faeces (Chi-Squared = 33.07, df = 2, p <>

Yes We Can

2009-10-27T14:11:00.000Z

Here's an old stencil made during the Obama Campaign. Yes, it's the Master Chief from Halo. I suppose I could change it now to "Yes They Did".

Lingua Tecnologia

2009-10-26T15:24:00.000Z

People worry that Capitalism and Technology will force everyone into a monoculture, destroying the idiosyncrasies of local cultures and languages. Here, I argue that, although these factors my cause people's cultures to grow increasingly closer, people's languages may diverge. This argument will take us to the North-West coast of India, but first, to a not-so-distant future.

Famers in Gujarat, India, using the Avaaj Otalo service.

In Jim Hurford's mini Sci-Fi Desperanto is set in a future society where people use lators - small devices which translate speech from one language into another - to speak to people outside their social group. Over time, language packs have been added so that anyone can speak to anyone else through them, even between dialects. This means that the languages of each little community are cut off from each other, leaving them free from the pressure to conform. The result is an explosion of diversity in languages.

The lators are used for trade between groups. However, similar systems may not be so futuristic.

Alastair Sussock was telling me about his Weather Insurance project for the IFMR Centre for Microfinance. It's an assessment of how farmers in the state of Gujarat, India, approach the idea of buying insurance against crop failure due to bad weather. Gujarat has an unstable climate where years of good rainfall will be punctuated by severe droughts, and weather insurance would help spread this risk. However, farmers were initially resistant to the idea of paying for a service which may not materialise. In the report, Sussock identifies 'financial literacy' as a major pitfall for the sale of insurance.

Indeed, his next project looked at how a mobile SMS system for disseminating local economic information would affect the farmers. There is a high degree of inefficiency in local markets, due to the farmers not being sure what price they will receive for their goods in which market towns. SMS farming is a system whereby farmers receive information about the prices of goods in various locations (e.g. 60 Rupies for 10kg of Cabbage in Ahmedabad) so they can best judge where to sell it and for how much. There are already several services set up (e.g. here), and many seem to be succeeding (here).

The idea is that farmers can optimise their business decisions, increasing profit and lowering prices. Armed with this financial knowledge, farmers are less dependent on social networks for information.

Daniel Nettle identifies ecological risk as a major factor in linguistic diversity. Where the environmental conditions are more variable and self-sufficiency is more difficult (e.g. a dessert/floodplain), groups must form close social bonds with each other in order to spread the risk of ruin. In this situation, it's more likely that linguistic norms are adopted by networked groups and, over time, everybody ends up speaking the same language. On the other hand, in ecologies where self-sufficiency is easy, there is less need to form close social bonds with other groups, therefore the languages will remain separate. Indeed, Nettle (1999) finds direct correlations between ecological risk and linguistic diversity.

Importantly, languages will only converge when groups have close social bonds with other, distant groups. Although the SMS technology may be expanding people's social networks, the new links are domain-specific and loose. Therefore, the introduction of ecological insurance and a centralised technical lingo for economics may have interesting effects on the linguistic diversity of India. If the ecological risk can be reduced by insurance, and the inter-group dependence and linguistic communication reduced through the SMS service, then the languages of individual groups will become increasingly isolated. This may, as Nettle and Hurford predict, lead to a diversification of language.

Although farmers across India may come to share a common financial language, the languages and dialects they use with those in their closest social circles will remain their own.

Single Couple

2009-10-26T10:22:00.000Z

Cultural Tansmission and Martial Arts

2009-10-25T13:54:00.000Z

As I tried to think of anything else except the searing pain in my quads at my last Kung Fu class, I started wondering about how martial arts are taught. The master gradually imparts a very stratified sequence of movements, called the form, to a student who then repeats them over and over. Importantly, the master tries to minimise the idiocyncracies in the student's form - a slightly arched back, irregular timing, poor balance etc. Because of this, I was wondering how much variation there is between different schools paracticing the same forms. Do different masters have different 'accents'?

The approach of most martial arts is specifically set up in order to minimise this. Eugen Herrigel, a german proffesor of philosophy living in Japan, describes his attempt to master the art of Archery. He claimed that, although the initial repetition of movements isolated from application was frustrating, he eventually experienced a 'direct transference of the spirit' from his master (from Singleton, 1998). How about that for anyone doubting the fidelity of memetic transfer?

The theory of evolution states that, in order for a system to improve, there must be variation and competition. In martial arts there is literal competition between sparring partners and in organised events. The best sequences will be selected. However, the system of teaching is set up so only the highest ranking teachers can alter and add to the form. This seems sensible, since changes are likely to improve the form, but it sacrificies improvements that could be found by chance.

Can we learn anything about the cultural transmission of language by studying the way form is transferred from master to student? Moscato (1989) in one of the first articles on memetic algorithms argues that Martial Arts is indeed a good candidate for the study of complex adaptive systems:

The form, like a chromosome, is not an indivisible entity. It is composed of a sequence of defensive and aggressive sub-units which can also be divided, a pattern that resembles the structure of chromosomes, genes and alleles. But within the form there are some movements which can be understood as an indivisible unit, and these are the ones that are really important. The whole is a support to let the brain transform them as reflexes that can be automatically triggered in real combat. The individuals can compute their fitness function by the evaluation of their performance in the execution of the movements of the forms and with some tournaments where they compete.

Moscato points out that most primates, including unskilled humans, tend to fight in an unorganised way while a martial art teaches one to use simple, efficient movements. This might be compared to the chimpanzee Nim's ramblings 'give orange me give eat orange me eat orange give me eat orange give me you' compared to a human's utterance ('give me the orange'). The martial art from may have a 'syntax' in the sense that the order of moves is very important.

However, there are several problems with using martial arts to study language. First, an estimation of the amount of variation in martial arts forms would be necessary. In lieu of excellent historical records, the best estimation would be reached by a comparative approach. This would involve the codification of forms then comparisons to find underlying structures. In other words, the whole problem of linguistics in the first place, and not obviously easier than for languages.

Secondly, it's not clear that martial arts has a semantics. In fact, the moves are not signals used for communication, but coercive actions.

Ah well, back to the searing leg pain.

Pun of the day #5

2009-10-25T09:00:00.000Z

E. - He didn't get my reference.
Me - To what?
E. - The Great Gatsby
Me - And now you have West Egg on your face?

Book Trailers

2009-10-24T16:55:00.001+01:00

I recently read Neal Stephenson's Anathem. Amazing, just like his Baroque Cycle, Cryptonomicon and Snowcrash. Although I'd love to see a movie adaptation, I realise that they are probably too complicated and too niche to ever get to Hollywood. Then, I found a trailer for Anathem online. But it wasn't for a movie, it was for the book:

I'd never heard of book trailers before. Seeing it after reading it was weird - some characters were different than I had imagined, but the re-construction of the fight scene is done word-for-word. I'm not sure how I feel about this. It had fairly high production values, and it was quite cool to see how other people visualised the scences. However, there's always the danger of being dissapointed by another person's idea of the book. For example, although I loved Phillip Pullman's His Dark Materials trilogy, I didn't much like the look and feel of the film.

However, at least with a film converstion, you have a chance to read the book first. Could book trailers start limiting our imaginaitons? Am I sounding a bit like an old person? Maybe. I started looking for other book trailers.

After hearing about Pride and Prejudice and Zombies, I was thrilled to see a well-polished trailer for Sense and Sensibility and Sea Monsters:

Although I couldn't find any other high-quality productions, I quite enjoyed the trailer for the children's book How to Save Your Tail, a re-telling of the 1001 Arabian Nights. It actually made me want to read the book. Brad Meltzer's trailer was also good fun, and looked like it was just made for fun too.

However, most of them were just text, pictures and a voice over (lamented by Phyllis Miller). Some have the tell-tale Windows Moviemaker blue background, others are a bit more snazzy, but ultimately are no more informative than a back-cover blurb. I found this one particularly annoying.

It turns out that book trailers are not a new thing. Here's an article from 2006 which discusses the difficulty of presenting images without influencing the way people percieve their characters. In a more recent post, Jonathan Fields argues that book trailers don't work because marketing advisors don't understand that a straight-up commercial approach won't get a trailer to go viral.

Actually, I quite like the idea, and hope it catches on. I'll be keeping an eye out for book trailers in the future.

Second Language Acquisition and Memory

2009-10-24T09:00:00.005+01:00

Evolutionary approaches have recently been adopted to try to explain the differences in adult and child second language acquisition (SLA). Hagen (2008) uses Paradis's (2004) theory of implicit and explicit learning to argue that adult SLA was selected out because it was not useful to our early ancestors. However, this order of evolution seems unlikely and a more parsimonious and longer-term explanation can be achieved by looking at the differing social pressures of adults and children.

Children acquire languages rapidly, effortlessly, implicitly and fluently. Adults, on the other hand, find second language acquisition laborious, difficult, and often do not achieve native fluency. Hagen (2008) addresses this difference from an evolutionary perspective. Firstly, Hagen shows that the difference between adult and child SLA is physical. Evidence comes from studies of SLA, and anthropology, which shows that the language faculty co-evolved with the physical evolution of our species. This part is well argued, and contains a diverse sources which, rightly, place the problem of the critical period as an evolutionary adaptation to the environmental conditions of the ancestors of humans.

Hagen cites Paradis’ (2004) theory of implicit and explicit knowledge being linked to procedural and declarative memory respectively. Procedural memory stores knowledge of unconscious, automatic skills such as walking. Declarative memory stores knowledge that can be consciously accessed, but with greater difficulty. The ability to incorporate knowledge into procedural memory atrophies in adults, which explains differences in loci of adult second languages in the brain. Hagen’s explanation for Bilingualism, however, focuses on relatively recent, certainly post-linguistic, environmental pressures. The main section of his argument considers the pros and cons of the ability for adult SLA. For the negative effects, Hagen cites Dyson’s (1979) argument that language diversity evolved to establish and maintain cultural differences between competing groups. This resonates with Accommodation theory. The benefits are mainly restricted to the ability for trade.

Hagen argues, however, that adult SLA would generally not be advantageous in pre-agricultural society (low populations dispersed widely). He links the critical period with studies of predatory animals brought up in captivity who don’t develop predatory behavior. As Hagen puts it, “If you don’t acquire it early, you might as well not acquire it at all” (p. 58). Hagen concludes that, since violent conquest of another culture yields greater rewards than cooperation during lean times, there was no pressure to develop adult SLA. On the other hand, the ability to socialize is vital to survival in a human society, so child acquisition is selected for.

Hagen’s explanation comes down to a discussion of the nature of man. Hagen argues that evolution, and evidence from history, favours an aggressive nature rather than an egalitarian one. Learning an enemy’s language is more costly and less beneficial than conquering them.

There are problems with Hagen’s approach, however. Although Hagen stresses that the differences between adult and child acquisition are physical, the explanation relies on sociological factors, originating less than 80,000 years ago. That is, although human language acquisition behaviour seems to “fit rather nicely into the mosaic of evolutionary theory”, Hagen has not explained the pressures that would cause the structural changes in the brain that would lead to differentiation in the first place. In the first half of the paper, Hagen suggests that adults lose their ability to acquire languages for two reasons: Firstly due to general cognitive aging. Secondly, because the language faculty is built upon the same instincts that lead to any young animal learning basic survival skills such as walking, and has therefore never been useful at later stages.

Part of the problem is that Hagen seems to be suggesting that SLA was selected against in adults (noted by Hirchfield, 2008, with a rebuttal by Hagen). That is, there was a point when language learning was equal for adults and children, and the adult ability eroded by drift. If this original adult ability was identical to the child ability (that is, there was no change over a lifetime), this seems to run against the idea that child and adult acquisition are physically different. An alternative situation is where adult and child learning was originally different, but achieved the same goal. This seems an inefficient solution.

Rather than a loss of an ability, perhaps the structural differences between adult and child acquisition reflect adaptations to different tasks. MacWhinney (2007, p. 2) argues the following: Children need to learn language for socialising in order to obtain protection and to inherit knowledge about how to survive. Adults face a different task – they must socialise in order to secure mates and provide for their children. Whereas there would be a pressure for children to acquire basic knowledge about the world (how to walk, red berries are poisonous) and how to communicate with their group, there would be a pressure on adults to keep track of social relationships (e.g., who is sleeping with who) and communicate about resources (e.g., where the food is, organisation of hunting).

I argue that this difference in social pressures is supported by two different kinds of memory: semantic memory, which encodes facts separated from context (e.g. the sky is blue), and episodic memory, which also encodes temporal information (e.g. it rained yesterday). Semantic memory has been argued to be related to procedural memory rather than declarative memory (Hurford, Flaherty & Argyropoulos, 2007). At any rate, it is certainly evolutionarily older than episodic memory, if not unique to humans, and develops later in childhood.

Adults can use the contextual informtion of episodic memory to keep track of complex social interactions. For example, it may be a bad idea to have a general sense that ‘X is a nice guy’, if X is nice only two thirds of the time. However, rather than an erosion of procedural/semantic memory in adults, it is the development of declarative/episodic memory that may be overriding procedural/semantic memory during adult second language acquisition.

I propose that the balance between using semantic and episodic memory was selected for so that semantic memory was favoured in childhood and episodic memory was favoured in adulthood. To test this assumption, a model was run to examine the ratios of semantic and episodic memory in a population of agents who age and compete for mating opportunities by assessing the social interactions of others.

The Model
In this model, agents can interact and store knowledge about events in either a procedural or declarative mode. Adults must keep track of how others behave in order to maximise their chances of offspring. It is hypothesised that this pressure will cause adults to use their declarative memory more than children.

Agents
Each agent is either male or female and has an age and an energy level. Each also has a set of genes that define a linear model for making decisions. That is, each variable involved in a decision can be modified by a slope and intercept with which to modify it. Agents receive energy at each time step (i.e. they eat) as long as they are above a certain age (they are initially dependent on others). Agents become capable of reproduction after a certain age with a small random variation. Agents can die either because they run out of energy or from old age.

Knowledge
Each agent has a declarative and procedural memory store which can store pieces of knowledge. A piece of knowledge can be either reflexive – another agent did something to me – or transitive – two other agents interacted with each other. The piece of knowledge also contains an attitude – whether the interaction was good or bad. For instance, a piece of knowledge can represent “Agent X gave me food”, or “Agent X and Agent Y mated”.

Memory
The declarative/episodic store is a LILO stack with a limited capacity. New pieces of knowledge are added to the start of the stack. This models episodic memory (e.g. “I gave Agent X some food earlier, but then they refused to mate with me”). The procedural/semantic memory stack works differently. Only one attitude can be stored with each interacting agent or pair of agents. New pieces of knowledge can gradually alter the attitude to another agent or interaction. When the capacity is reached, the least strongly held opinion is deleted. This models semantic memory (e.g. “Agent X is friendly”).

The Model
At each time step, the agent decides either to:
• Mate, based on age, energy and puberty
• Give food, based on age and energy
• Teach another agent, based on age and energy
• Observe another event, based on age and energy
There are a number of time steps in a year.

Giving
Agents can choose to give food to another agent. In doing so, their own energy depletes and the receiver’s energy increases. The receiver learns that the giver gave them food (a positive attitude).

Teaching
Agents can choose to transfer a random piece of knowledge to another agent. The receiver also learns that the teacher chose to teach them something.

Observing
Agents can choose to learn any event that occurred in the current time step.

Mating
If an agent decides to mate, they choose another agent of the opposite sex to mate with, based on their knowledge. For each agent they know about, an overall attitude score is calculated based on the sum of the attitudes attached to each piece of knowledge and a weighting gene. Once the potential mate is chosen, they attempt to mate. The reciever may choose to accept or decline, based on their own attitudes towards the initiator. If they accept, then both agents expend energy and, if the receiver has not expired, they become pregnant and a new agent is added at the end of the year. An agent cannot become pregnant with more than one child at a time.

Reproduction
Sexual reproduction is modeled with genes chosen randomly from two parents of different sexes. Each gene represents either a slope or intercept for a decision variable. There is a small chance of drift.

Model initiation
The model was initiated with 50 agents with random ages, energies, sexes and gene settings.

Preliminary Results
The graph below shows some preliminary results from one run of the model. The Yellow line shows the population size, which slowly increases (on right scale). The solid lines represent the percentage of the population choosing various activities (Red = Mate, Blue = Observe, Green = Teach, Brown = Give). The dotted lines show the average proportion of choices to use episodic memory to store new knowledge compared to semantic memory, for adults (black) and children (brown).

The graph shows that children's average use of episodic memory is lower than adults'. This suggests that, over time, agents evolve to use mainly procedural memory as children, but then start using more episodic knowledge when they become adults. The robustness of these results need to be examined further.

Discussion

The preliminary results agree with the hypothesis that the ability to store temporal social information using episodic memory during adulthood was selected. If the use of episodic memory conflicts or draws resources from semantic memory, this may explain the difficulty adults have in learning a second language.

This theory suggests a change over a much longer period than Hagen's and is also based on an adaptive advantage for declarative/episodic memory which is more likely to be selected than a neutral effect of procedural/semantic memory in adults.

The theory may be important in explaining linguistic diversity also. As Nettle (1999, p26) points out (under his 'Neutral model' of linguistic variation), in order for languages to change so as to become mutually unintelligible, two mechanisms are required. First is a source of variation. This is easy to find in both speech production by adults and learning patterns in children. The second type of mechanisms are amplifiers of variation - factors that maintain differences between two populations. This includes geographical isolation, but as Nettle points out, most societies have considerable levels of exposure to the languages of other cultures. This last point is very strong, but perhaps overlooks who is participating in cross-linguistic activities. If adults are the main interactors (trade, marriage, politics, war), then linguistic diversity may be maintained because they are poor at acquiring novel linguistic variants. That is, variation may be maintained if only the members who are acquiring the language (children) are isolated from other linguistic variants.

Since there is a huge amount of linguistic variation in the world, the selection process described in the model above should have occurred very early in language evolution, if not pre-linguistically.

L. Kirk Hagen (2008). The Bilingual Brain: Human Evolution and Second Language Acquisition Evolutionary Psychology, 6, 43-63

Immersive Worlds

2009-10-23T15:52:00.002+01:00

A Replicated Typo recently posted on how bad the new Sci-Fi series Defying Gravity is. Charles Stross's blog on approaches to Sci-Fi is cited, basically making the point that a lot of Sci-Fis add lazers to ordinary dramas without thinking about the consequences for the characters.

It's a good point that many sci-fi programs seem to have fancy technology just for looking at, without any of it changing the way the characters interact or think about the world. However, there are some TV Sci-Fi shows that build the world from the ground up. My favourite example is the Ghost in the Shell spin-off TV series Stand Alone Complex. The show follows a team of anti-terrorists in the year 2030, where society has changed in response to the invention of e-brains and cybernetics. I often got the feeling that Masamune Shirow did not create a world, just write about one he happened to visit at some point.

There are two things that make SAC feel like a total immersion in a different world. Firstly, because it's animated, there is no difference between 'real' and 'CGI'. That is, suspension of disbelief is easier. Secondly, the new technology has a direct affect on the lives of the characters. Not only does the proliferation of cybernetics have an effect on what it means to be human, there are new ways to interact, and new ways to break the law (mind-hacking). Similarly, Orwell's invention of 'Thought-Crime' in Nineteen-Eighty-Four created a culture with new pressures.

As a side-note, also similar to Nineteen-Eighty-Four, SAC looks at the balance between public privacy and public security. However, the heroes of SAC are constantly hindered by lack of surveillance, and the viewer comes to empathise with their need for more information, quite contrary to the reader's response to Winston Smith's situation.

Back to Defying Gravity, it's difficult to see how a small crew going on a single journey aroud the solar system would change a culture. Perhaps Defying Gravity will eventually get around to this. I don't doubt that space travel can change a society's perception of itself, but how different is it from the moon landings for the average person?

Here we get to the root of the distinction between immersive worlds and drama-driven worlds. In the drama-driven world, the big change usually only occurs for a small, elite group of people. For example, Defying Gravity. In an immersive world, the big difference is global and the protagonists are more ordinary people who must survive in it. For example, Nineteen-Eighty-Four or Ghost in the Shell.

Furthermore, the aims of both approaches now becomes apparent. Ordinary society + extraordinary people = drama. Ordinary people + extraordinary society = social commentary/satire. The examples continue: Heroes vs. Dawn of the Dead, Star-Trek vs. Neal Stephenson's Anathem, Dr. Who vs. Watchmen.

So, maybe it's just a lack of satire that A Recplicated Typo is missing. If so, you've got the formula now, so go out and make your own show.

Labyrinth

2009-10-23T12:37:00.001+01:00

For a while my life ambition was to be the keeper of a labyrinth and confuse people. And I'm not talking about walking labyrinths used for meditation. Not even proper mazes that are fantastically designed (such as Tom's Mazes). I'm not even talking about puzzles just put there to infuriate everybody, such as this disastrous idea. I'm talking about a proper labyrinth with trips, traps and treasure.

Here's an old plan for the entrance. You enter a very high courtyard and see a tower in front of you. You climb the stairs and find yourself looking down on the courtyard (the walls of the courtyard are taller than the tower). And that's it. The trick is to realise that the turns in the staircase are actually slightly less than 90 degrees, meaning that, by the time you reach the height of the top of the tower you are slightly twisted, allowing the exit of the stairs to come out in a different, identical courtyard. To get through the entrance, you have to climb down the outside of the tower and leave through what looks like the entrance.

I even went to the trouble of testing whether it was feasible using Unreal Tournament's Editor:

Sometimes I wonder why I didn't put any effort into cool things.

Anyone else want to buy an abandoned mine and build a labyrinth? Preferably someone who has experience in body-swerving health and safety requirements.

Hypocrisy

2009-10-22T09:30:00.001+01:00

Stung by Nettle

2009-10-21T17:37:00.000+01:00

I've been reading David Nettle's 'Linguistic Variation'. It brings up a very concerning issue:
A common ancestor to humans lived about 140,000 years ago.
Assuming a generation time of 25 years, that's 5,600 generations ago.
There are about 6,500 mutually unintelligible languages in the world.

At first glance, this looks like there is at least one new language invented every generation! However, once a language has split into two, both can then go on to split into two other languages each. This means that it only takes 12.66 changes to create 6,500 different languages, or one new language per group every 442 generations, (or 11,000 years). Of course, this does not include languages that were created then died out, but even so, it's a lot less impressive than at first glance. In fact, what this suggests is that language transmission is actually pretty robust in the sense that it takes a myriadum (perhaps someone with better latincan correct me on this) for two sub-groups to develop so they can no longer understand each other. This agrees with Pagel, Atkinson & Meade's (2007) estimations of rates of change of single words over roughly the same time period.

Moonlighting

2009-10-21T11:10:00.000+01:00

Van Morrison's 'Moonlight' is considered one of the sexiest songs ever, but I always doubted the lines:

"Well I want to make love to you tonight,
And I can't wait 'till the morning has come"

What's the hurry, Van? I thought you wanted to dance with me but are you, in fact, terribly embarrassed to be seen with me? You just want a quick in-and-out, wham-bam-thank-you-ma'am? Why are all the men in my life like this? Damn you, Van, keep your heart-strings, I'm going to dance with someone who'll stick around.

Pun of the day #4

2009-10-20T14:23:00.002+01:00

Returning to work, knowing that a colleague with a sore throat would be there, punctuating the silence every three seconds with a hoarse croak:

K. - We'd better get back to the office.
Me - You mean the coffice?

Spike

2009-10-20T12:27:00.000+01:00

Spike from Cowboy Beebop

This hangs above my desk and reminds me that no one else is going to do my work for me. The image is taken pretty much as-is from the final shot of the final episode (here, with immense spoilers). It's one of the best endings of any TV show ever, with a beautifully subtle twist in the final seconds of the credits (noted also by BassInSpace), and this is my little tribute.

However, there's currently a live-action movie of Cowboy Beebop in production, with Keanu Reeves starring as Spike. So now I have a picture of Keanu Reeves on my wall. Not exactly something I had planned.

2009-10-19T17:30:00.003+01:00

I'm thinking about running an experiment into a genetic basis for the inheritance of cultural traits:

Hypothesis:
Cultural traits (fashion sense, taste in music etc.) are genetically determined, so that people with similar genes will tend to learn behaviours in the same way.

Method:
Get two identical twins.
Separate them until they both decide to get a hair-cut.
Record their haircut choice.

Prediction:
If one subject shaves off all his hair, the other will too.

I'm thinking about calling it the bald-twin effect.

Pun of the day #3

2009-10-19T15:12:00.001+01:00

E - Oh, I wanted starters.
Me - You don't hors d'œserve them.