Wednesday, 22 May 2013

Method Madness

I recently found this passage in doing some background reading on narrative.  It's the kind of excerpt that makes me want to get up and do a cartwheel or yell jackpot! and confirm to everyone sitting nearby that I've finally lost it.  (If you don't have the same reaction, that's probably good.)
"Sir Frederic Bartlett, a celebrated Cambridge psychologist, was the first scholar to investigate and theorize cross-linguistic and cross-cultural differences in narrative construction.  In Bartlett’s (1932) classic study, Western subjects were read a Native American story, The War of Ghosts, and then were asked to re-tell it.  Because the participants found both the story structure and many accompanying details unfamiliar, they repeatedly transformed the tale in recall, both through omissions of details and through rationalizations, which made the story conform to a more familiar Western pattern.  On the basis of these observations and experiments, Bartlett (1932) developed his theory of schema that informs much of contemporary cognitive science, psychology, and narrative study.(Pavelenko, in press, p.7)
What I find so exciting about Bartlett's experiment is that it is so similar to my own, but it's from 1932!  It tests the culturally defined pattern of narrative which is linked to how we make sense of the entire plot line of the story.  Discovering that my methodology has some precedent which I have now adapted and appied to communication via ICT strengthens the validity of my approach.

Here in Gulu, Uganda, I am investigating the reverse phenomenon as Bartlett.  I'm interested in individuals far from the cultural of design...telling their own stories through a western artifact.  (Some Acholi speakers may call phones and tech related objects "things for work made with craftsmanship in iron" or alternatively, "things for work made with with skill from whites")  I hypothesize that when a narrative is given in Acholi via an ICT, a narrative shift occurs, but it is the ICT which triggers this shift.  The shift is between an Acholi narrative structure and a western one, and the spatial cues of the technology 'space' we enter when using ICT applications causes users to adapt their narratives and concepts to fit the western model much in the way a bilingual Acholi-English speaker makes small changes when switching into English.  This change happen at the cognitive level, the level of categorization, ordering, and many other 'thinking' level pre-language processes that the speaker may or may not be aware of.  My hypothesis is that the ICT, even if the interface language is in Acholi, is recognized to be 'of the west' because of other visual cues which every culture makes sense of in different ways. (Check out international signage for some great examples.)

My hypothesis is based on years of field observations in a range of linguistic and cultural settings, but the challenge is how to create a research method that captures a phenomenon I have a strong inkling about in valid and reproducible terms that the academic community will also find compelling.

Other research that I find cartwheel-worthy comes out of ALT-I, the African Languages Technology Initiative.  In particular, the engineers Odejobi and Adegbola theorize, 
"services supporting CMC [computer-mediated-communication] intended for use in African environment should exploit and implement language technologies developed around African languages and cultures."
 They propose this addition to current technologies of American and European origin should first,
"describe and represent the knowledge systems underlying African systems of communication in a form amenable to computation, e.g., numerical, graphical, or symbolically…. by critically and analytically address[ing] the question of how African people represent concepts." 
My research is therefore grounded in conceptual transfer theory within cognitive linguistics.  Their idea is both broad and ambitious, and my research begins to explore the possibilities they suggest.  If we concede that there are a myriad of ways in which different cultures communicate, why is there only one style of communication technology as research teams led by both hill Hill and also Zakaria propose, only the western-engineered model of sharing our narratives, transmitting stories, moderating the digital information that has become interconnected with our very identities?  This research examines the impact of information and communication technology design-- the current mono-cultural design-- on narrative, identity and participation with examples from a bi-lingual Acholi-English case study in Gulu, Uganda.

...and this case study involves what exactly?  That is what I have been explaining to community leaders for the past several weeks in order to get the OK to start collecting data. (My Acholi explanation is getting better slowly and involves the word apoka poka  which means difference)

Odejobi, T. and T. Adegbola. 2010. Computational and engineering issues in human computer interaction systems for supporting communication in African languages. In: O.A. Taiwo ed.  Handbook of research on discourse behavior and digital communication: language structures and social interaction. Chpt. 56. [ebook] ISBN: 9781615207732 [Accessed 20 January 2012].

Friday, 10 May 2013

For the Birds and Mr Brooks

(this is a break from Gulu and current research while I'm out in the field collecting data, but check out my piece with the Policy and Internet Blog out of the Oxford Internet Institute)

This is a response to many pieces I've read on big data analysis and large scale social media analysis and, in particular, a recent NYTs op-ed column by titled, 'What You'll Do Next' in which Mr. Brooks wrote:
"The theory of big data is to have no theory, at least about human nature. You just gather huge amounts of information, observe the patterns and estimate probabilities about how people will act in the future. . . .

To discern meaningful correlations from meaningless ones, you often have to rely on some causal hypothesis about what is leading to what. You wind up back in the land of human theorizing."
Mr. Brooks is correct, and this is something I've written about before.  As much as big data scientists hate to admit it, there is social science underpinning their algorithms and interpretations.  Teams at top institutes are using a theory called homophily.  And just hearing the name turns my stomach.  Not so much because of the word (although, it is an odd term), but for the same reasons I grimaced when I heard the book Three Cups of Tea spun as a handbook for foreign policy.  However, since my visceral response is not a recognized metric (yet), I will enumerate my objections to the theory’s current application to big data mining.  

It is a social network theory roughly asserting that it is our common ‘likes’ that bind us.  It has inspired research papers with titles such as, ‘Birds of a Feather [Stick Together],’ which chronicles many types of social network theory.  Did I mention, this is all sociology, not information science, and it’s the study of humans relating offline out in the world?  
Yes, these theories started back around the 1960s in the United States when the racial upheaval motivated social scientists to ask, ‘What is it that binds us together at all?’  The insights they gained from work stretching into the 1990s, perhaps combined with the familiar word network, has attracted researchers from information science to apply the finding to the online domain.  
Now assuming that humans behave offline in the same way they do online is one leap, but big data scientists have made yet another.  The heaps of data come from many sources such as social media, applications, and devices.  Headlines were made when researchers at MIT predicted the political leanings of mobile phone users and even tracked the spread of illness based on users' habits. The ground-breaking results from MIT were based on only American user behavior, but discussed as though they were universally applicable.  

Results assumed the one user/one device/one account rule of the US, but this isn’t the pattern in some communal cultures.  Social scientists are just beginning to study user engagement with technology in places like Nigeria and Indonesia and discover how much we thought we knew, how much we assumed was universal, does not hold up under scrutiny and is increasingly dynamic. 
Bulk and ease-of-access to data does not immediately add up to persuasive conclusions.  I am not convinced by the argument I hear so often from big data proponents, ‘the data speaks for itself,’ because the models I come across that are essential for any human to make sense of the tonnage of data are based on cold war era foundations.  Theories like homophily do not take into account the advent of the internet or cultural variations. When interpreting social media data or making assumptions about user behavior, culture is a variable that cannot be ignored. 
The truth is we don’t know what to do with all this data yet.  And I am a bit torn here because my inner-engineer wants to build better models, to improve.  I am fascinated by the problem of how to incorporate cultural variation and increase what we can learn from the rich amount of information at our disposal.  But what will we use this be used for?  Researchers at Harvard's Berkman Center aspired, through this flawed method, to create a model of the Iranian blogosphere as 'unique as a snowflake.'  I probably don't need to explain the value of this research, but the social science foundation proposed simultaneously that all humans behave in a similar and predictable manner and also that unique cultural insights can be gained from a model that ignores cultural variation. (If you got lost in that last sentence, you're actually right where you should be. Most research grounded in homophily makes about that much sense.)  So this is where I hope the larger community of scientists, social and data scientists, can have a rigorous debate about how to do better... make better models and concern ourselves with the broader ethical implications. 

Wednesday, 1 May 2013

Like it's 1899


        When we talk of bridging the gap, the technology gap, the ubiquitous construction metaphors make it seem as though we are in the Age of Victoria building train tracks across every known expanse and into the depths of continents, clearing the way for the new age, for industry, speed, access to modernity.  Every aspect of this approach to getting people online and engaged with technology should come with a top hat and monocle because it’s right out of the 19th century.  Most solutions to access focus on translation, getting interface scripts or keyboards to register languages besides English.  This seems like a job for a linguist in 1899. . . .To go off and chart, transcribe, and translate foreign languages so they could become part of the world of knowledge.  (That world has changed.) As if nothing exists, can truly progress, get organized, move forward, until it is written down.

Software engineers tell me anything with software is possible.  So let’s leave the horse and buggy behind.  Divorce ourselves from writing.  Throw out the keyboard.  Lose the map.  Unlock the individual user password profile.  There is another way.

Every engineer I tell this to says, 'Fine, so now what?'  Well, I am still working on that part, but essentially, it's up to indigenous (that is to say local) developers to create what makes sense where they are.  My research is an experiment which will quantify a few points where the current ICT design is 'pinching' the most.  Where it is preventing oral cultures from sharing narratives, from organizing information, from conveying concepts in the way which seems most natural in their thoughts.  (The thought/pre-language moment is contested and debated within philosophy, psychology, and linguistics.  It defies definition.)  The areas where I have seen the current ICTs 'pinch' and where I think we should be looking to make some changes are as follows:

3 Guiding Points to shift out of chirographic and into oral cultural design mode:

1. Information is gathered, collected, and moved.  But only what we can carry, in sound and echo.  It moves through repetition.  It is stored in our minds and in our retelling, over and over in the sound our retelling makes.  It does not have the permanence of writing, the ownership of authorship.  It forms by accumulation, being born from many sources; it is fluid, varied from omissions and additions.  Recursive.  How is this different from US culture?  It’s something to do with how we experience time.  And something to do with maps.

2. Group.  There wasn’t a word for group in Acholi until it was borrowed from English.  Gathering, yes, such as a formal occasion like a wedding.  But group, no.  My contribution in the Group project. Women’s Group.  Book Group.  He left the Group to become a Solo artist.  Groupthink.  In English, we call attention to the group phenomenon, the crowd.  But for a communal culture, the group is the norm.  As the religious scholar John Mbiti explained using Descartes’ terms:  
“I am because we are, we are therefore I am.”   
There is a word for alone, the exception.  And technology is a tool of isolating, individualizing power.  S. Turkle writes extensively about this phenomenon at MITs Initiative on Technology and Self.  The question of how a tool that is inherently isolating will impact a fundamentally communal culture should be as much of a design consideration as the interface language. 

3. There is no word for map in Acholi or any of the languages in the region.  It was borrowed from English, but seems to carry the meaning: list, unnecessary things white people must write down to read, e.g., menu, see ‘vegetarian map.’

More about maps...  First, in Fez, among ethnically Tashelit (Berber) friends speaking Moroccan Arabic driving looking for a cousin’s new house where none of us had previously visited.  In the US, if I were on my way somewhere for the first time, I would perhaps have a gps system in my car, get written directions from the person I was visiting, or consult the internet.  I would also consult a map and probably print all of these things or have them available on my phone.  Not in Fez.  There, in the back seat,  I listened to the driver call ahead and ask where to turn then hang up.  We drove a little way, came to the roundabout, made the turn, then pulled over.   Another phone call.  Another instruction.  Singular.  Not a series.  The phone was not passed to a passenger who could write things down nor were several directions said out loud so we could all help remember.  This pattern of calling for one step, one instruction at a time continued turn by turn, landmark by landmark, through the old city and out into a suburb until we reached our destination. The cousin talked us there.  Step by step.  In real time.  (Isn't there a device you can buy that does that?  What's the difference?  See #1 and 2)

Again in Gulu, Uganda in Acholi.  Because there are no maps and not so many signs (and the ones that exist don’t reflect the information you hope for) you have to ask directions... frequently.  When you do, the most common answer, in English is,  ‘Just go along and then you’ll see it there.’  Or, with an arm outstretched, pointing at four buildings, ‘It's that building there.’   So I was curious if there was some part of 'there' which didn't translate from Acholi, maybe there was a tone which imparted more information.  And upon comparison, there is certainly more than one 'there,' but also it's connected to memory and experience.  Extensive research describes how bi-/multilinguals remember things differently in different languages.   And this would be an entire project in itself, but my impression is that the landscape and experiential memory, the sensory data we collect to know a place, to create detailed mental maps, this has been created in Acholi not English.  So if you get directions in Acholi, there are many, many details.
"The white building by the tall tree next to the butcher, be careful of the road there it's not been repaired... and not far from the little road you take to get to the flower shop, have you been there?  It's very nice my daughter works there."  
But to get back to designing ICTs, when a user must shift into English mode, either through interface language or through other visual cues which say to the user unconsciously, 'this is an English space, go into English-mode in your brain,' then their ability to convey details about their environment diminishes if they are shifting out of Acholi and into English, even unconsciously, at the conceptual level.
These 3 points-- information flow, self/group, and maps-- illustrate what I will write more about in later posts and what I hope to bring to light with my experiment here in Gulu, the interconnectedness of time, space and language.  So far, I am pleased to report that these ideas seem to resonate with people here in policy-making, in ICT, and basic users of the current tech. (To my surprise in fact, my research seems to be better understood here than in the US/EU, almost as though I was explaining something very obvious that everyone here already knew.)