My life in Abancay, Peru

Abancay, Peru
Friday, July 28, 2006

I have been so busy lately, that I haven’t had much time to sit down and reflect on what I’m been doing here in Peru. Was it Descartes who said that the unreflective life is the life not lived? By that measure I haven’t been doing much living lately.

Currently I’m in Abancay, a plain little town in a cleft between hills which somehow grew into a provincial capital despite itself. I’ve never been in a Latin American provincial capital which didn’t have a plaza, but somehow Abancay grew without town center surrounded by the requisite Cathedral, ayuntamiento (town hall) and other accoutrements of colonial rule. I can’t say that there is much to see here in Abancay, but that is part of the charm of the place. The tourists don’t stop much, except to catch the next bus to Cuzco or to head off to the hills on one of the scenic hikes. In other words, there aren’t too many gringos hanging around, to make people stop being curious about why a strange gringo would show up in their town.

The reality, is that I never would have chosen to spend much time in Abancay, but Daniel Yucra, a professor of computer science at UNAMBA (National University of Micaela Bastidas of Apurimac) invited me to come. Since UNAMBA is a new university and all the professors are recent arrivals themselves, the institution is still seeking to define itself. The computer science department is pretty s basic by most US university standards. I get the feeling that I know as much about programming as some of the professors. They teach PHP and Pascal for Delphi. It’s not exactly high powered programming, but they are still defining their list of courses in the university and seem open to listening to the ideas of the crazy gringo. Daniel has been very gracious and has taken me to see everyone he can think of in Abancay. Somehow I get the feeling that I am the strange zoological exhibit that he loves to show off.

At any rate, my appearance is a sort of validation of the importance of free software. The ability to attract foreigners really highlights the importance of his work with free software. Daniel is one of the founders of and the Tumix Linux distribution and was invited to come and teach at UNAMBA partly because of his expertise in Linux and free software. Daniel has been taking me to meet all sorts of people to present our Quechua software project.

I was pretty apprehensive about my ability to navigate socially in Peru. One thing about Peru is that it is very important to establish your class credentials by the clothes that you wear. In the US where everyone has washing machines and the money to buy a suit, clothes have become largely meaningless as a marker of status and position in society, but not so in the Andes. Professors are professionals and often wear suits, or at least ties. There is a little bit of leeway for foreigners, but you still have to fulfill certain social expectations. I got off the bus in Abancay in my hiking boots and jeans, and headed for my hotel to rest after a sleepless night in the bus. I first stopped in the nearest cyber-cafe and sent Daniel a quick note, telling him that I had to sleep, but we could meet the next day. But he was so excited to meet me that he immediately showed up at my hotel and took me to the nicest restaurant for lunch. Afterwards, he took me to meet everyone he could think of. I was kind of embarrassed considering the fact that I was half asleep the whole time and I must have looked horrible after 14 hours on the bus without even having the chance to change out of my grungy hiking boots and jeans.

I redeemed myself by showing up a UNAMBA for the next four or five day in slacks, dress shoes and and a nice button down shirt. Daniel kept ushering me into offices to present our Quechua software project. After 5 days of wearing the same set of clothes, I thought that I was safe and had meet everyone that I needed to meet, so I showed up in my typical jeans. It was on that day that Daniel decided to usher me into the office of the president of the university for an hour-long meeting to present the idea that the university should sponsor and support our project. Despite my lack of social etiquette, the president seemed very receptive to the idea.

I discovered that I have as much credentials as most of the professors at UNAMBA. Very few of them have doctorates and some only have the equivalent of bachelors degrees. Nonetheless, I was shocked when professors started inviting me to come to their classes and speak for an hour about our project. After talking to 5 or 6 different classes I seemed to have gotten my spiel down pretty well. The professors seem perfectly happy to let a strange gringo show up and rant about the important of free software, the disappearance of indigenous languages, and how we need to become active co-creators rather than passive consumers of technology if we don’t want our thoughts and culture to be controlled and defined by transnational media and technology corporations. Somehow talking about software in Quechua seems to valid course material for any class. I have talked to computer science, education, veterinarian science, and even prerequisite filler classes. They had me get up and talk at a computer science conference for 50 minutes. I can’t imagine a foreigner showing up in a university in the US and getting the same invitations to speak to so many different groups. It is hard to talk to people about technology, intellectual property laws and the importance of the free exchange of ideas, media and technology in a foreign language, but I have always been good at talking about things that I care passionately about.

Whenever I meet people in Latin America, I have to repeat my name several times, and only seem to satisfy them when I tell them that “Amos” was a prophet in the bible. Maybe I was well-named because I do find myself preaching a message of imminent doom and hope for salvation. The doomsday message is that Peru is in the process of signing a free trade agreement with the US and will have to abide by the intellectual property laws of the US. I warn people that the Business Software Alliance will show up in Peru just like it showed up in Guatemala 5 years ago and start busting cyber-cafes and schools for not having legal versions of MS Office installed on their computers. Considering the fact that the annual license in Peru for MS Windows is over $200 and MS Office is close to $300, educational institutions in Peru really do have a reason to be alarmed. The president of UNAMBA seemed to be listening when I told him that my university in the US has to pay Microsoft 7 million dollars per year. That is probably more than the entire UNAMBA annual budget.

When I haven’t been talking to people, I’ve been desperately trying to finish my PHP script. What started out as a simple little program to grab phrases from one translation file and insert them into another translation file has grown to almost 3000 lines of tangled code. It is really ugly coding because it didn’t have a clean design from the get-go, but it works and that is the important thing. I managed improve its execution time on large files by a factor of 5 by indexing the phrases for faster searching.

This one project in PHP has convinced me that I will never use the language again if I can help it. I started programming in PHP because I wanted to create a simple web tool, but my script hogs so much processing time that it would slow my hosting server to crawl, so I decided to release it as a tool to be downloaded and run on a user’s machine. Not only is PHP slow, but it is a downright nightmare if you want use it as a command line scripting language. I’ve found the language to be more hassle than its reputed worth. The fact that I have discovered two bugs in the language certainly doesn’t increase my estimation of PHP. Unlike other free software scripting languages like Perl, Python, and Ruby, PHP simply wasn’t designed for writing multi-lingual, multi-platform command line tools. Unfortunately, I’ve put too much work into my program to easily switch to a better language now. PHP doesn’t have any way to query MS-Windows to find out what is the system language, so I had to resort to searching around in the file system and looking at standard Window’s directories to figure out whether to put the interface in Spanish, English, or Portuguese. In MS-Windows PHP uses archaic DOS rather than creating its own command line window like Python does. This means that all output has to be converted into crazy DOS character sets like CP1250, but again PHP doesn’t provide any way to know what character set to use. So if a user’s machine doesn’t like the character set that I chose, it is going to display a lot of garbage on the screen. PHP string functions don’t even fully support UTF-8, the modern standard for representing all languages with the same character set. Even more aggravating is the fact that PHP doesn’t have built in support for gettext so you have to hope that gettext and intlib (standard libraries for software translation and internationalization) are installed on the user’s machine. A real language like Python has these things built-in so you can safely run your code anywhere in the world. I had to implement my own jerry-rigged gettext function so people could use my program in something other than English.

I keep banging away at my PHP script because I have a problem which nobody else in the world has bothered to solve. Most translation of free software happens in parts of the world where there is a sufficient number of people who can read the original English version of the program in order to translate it into their native language. The problem is to there are very few Quechua-speakers who know how to read English, so they need to see the Spanish translation in order to translate into Quechua. None of the existing translation tools allow for trilingual translations in this manner. I noticed that there are almost no translations of indigenous languages in parts of the world where the major written language is not English. For instance, there are a number of translation projects in Africa, but they are all happening in the African countries where English is the major written language such as South Africa, Zambia, Swahili. There is little free software translation into indigenous languages in Latin America and French and Portuguese speaking Africa. To fill this gap, I decided to write a tool to insert translations from a 3rd language into the translation file so people could translate from a language other than Engish. I hope that it will make translations possible in the non-English speaking world. If you find this at all interesting, you can read the help file at or in English at and download the program at

Right now I’ve got English and Spanish versions of the program, plus my mangled Portuguese translation. I had to do it without benefit of a Portuguese dictionary. I was shocked that none of the faculty here at UNAMBA have a Portuguese dictionary that I could borrow. Of course the joke of a library at UNAMBA doesn’t have a Portuguese dictionary. Almost 70% of the people who live in the Apurimac region speak Quechua, but the university’s library doesn’t even have a Quechua dictionary–in fact it only has one book about Quechua and it’s mostly in Spanish. I have as many books about Quechua as some of the Quechua professors at UNAMBA. When I showed up, they all wanted to copy my Bolivian Quechua books, because they had never seen them before. Shara Huaman, one of the Quechua professors, even asked me if she could copy my illegal copy of the dictionary from the Academia Mayor de la Lengua Quechua in Cuzco and she is a graduate of the Academia Mayor. She told me that she had been searching for the book for a number of years, but had been unable to find it. It’s the best Quechua dictionary yet produced, but only 2000 copies were printed in 1995, so it is almost impossible to find today.

I’m constantly amazed at the sparsity of decent books in the Andes. In the US, I have access to more books about the Andes, than someone living in the Andes, including many of the books published in the Andes. Through inter-library loan, I can get my hands on almost any book in the world, but Latin Americans have to depend on having their own personal libraries or a network of friends and colleagues who can pass the books down the grapevine to them. Because books and access to information is so restricted for many people living the global south, the internet becomes doubly important as a means to spread information in places where people don’t have access to decent libraries.

It is in this context that the idea of creating dictionaries online becomes so vital. I’ve been kicking around the idea of trying to create a dictionary online in the same way that wikipedia has created an online encyclopedia. Because there are thousands languages in the world and many regional variations, there ought to be an easy way for people to go to the internet and record the words that they use in their region without deferring to the so-called experts. It becomes a living document, under constant revision and review as everyone has the right to add to and edit the dictionary. In the end, the technology is democratized as the final product, like wikipedia, is often more comprehensive than what the “experts” can create. In the past, the foreign linguists or evangelical missionaries were the ones soften defining the alphabets and dictionaries for many minority and indigenous languages, but the speakers themselves should have the ability to define and register the written form of their own language. You only realize how important it is to give common people the tools to record their language when you realize that there were roughly 30,000 languages in the world, but only 6000 of those languages still exist today. At the rate that they are disappearing, there will probably be only 900 within a century (if I recall the statistics correctly from Larry Diamond’s _Third Chimpanzee_). It sounds like a simple proposition to democratize the tools to create dictionaries, until you realize how that there is no consensus about the proper written form for most oral languages. For instance, in Quechua do you use 3 or 5 vowels? Do you use H or J to represent the sound that is h in English, but j in Spanish? When you have 6 |k| sounds (k, k’, kh, q, q’, qh) which are often confused and often vary regionally, which is the correct way to write a word? The experts can’t even agree, and most quechua-speakers have little idea how to read or write their own language, except to sound it out in the Spanish alphabet. At the end of the day, whoever is adding to the online dictionary has to have some “expert” knowledge. Still I have the dream of creating an online dictionwiki.

Of the roughly 35 indigenous languages in Peru and 25 languages in Bolivia, probably only Quechua, Aymara, and Guarani in Bolivia will survive the coming century and those that do survive will have to adapt to new contexts, especially new technologies. So far Quechua has made little transition into what is considered the things of modernity. I was fascinated to see how a musical group like Uchpa sings in Quechu accompanied by electric guitars and a rock and blues beat. Unfortunately, this sort of transition into “modernity” is usually accompanied by the abandonment of the language. For this reason we are trying to carry Quechua into the context of computers to show that the language can make that transition into “modernity” as a living and adaptable tongue. It is fascinating how even people who speak Quechua seem to operate under the built in assumption that Quechua is simply the language of the countryside. When I tell them about our project, they say, “Oh! This will really help those people who don’t speak Spanish to use the computer.” The percentage of monolingual Quechua speakers is roughly a 1/4th of speakers and it is diminishing every year. But very few of the people who only speak Quechua will every touch a computer in their lifetime. In short, they have instantly relegated our project to the use of a couple thousand people. I tell people that our project is actually aimed at the youth in the city who are most in danger of loosing the language, because they don’t think it is useful or “cool” in the modern context. They are the ones who are mostly to be using a computer in the first place.

If I were an anthropologist, I would simply record people’s opinions and write a lovely essay describing people’s attitudes about the contextualization of Quechua, but I decided that I am more of an activist than an academic. I want to change the world more than I want to study it. So I sat down and wrote a manifesto calling for volunteers to help us adapt Quechua to “modernity” as part of a free software project. I was a little apprehensive when I took it to Apolinario Saldívar, one of the Quechua lingüists at UNAMBA, but he instantly grasped what I was trying to say and helped correct my grammatically- challenged Spanish. I was rather pleased that he agreed so readily with the general principals of the manifesto and only offered a few minor changes. Daniel Yucra also agreed with the declaration, which is hardly surprising considering his background in free software. I haven’t had a chance to get Manuel Ibarra’s opinion and in many ways he is the most critical voice, because he is both a Quechua-speaker and a professor of computer science, so he understands the two worlds that we are trying to merge better than the rest of us. If interested in our project, our current draft is posted at or in English at for review. We are trying to get it translated into Quechua before we send it out to the world. We want our title to be “Free software for everyone”. I had no trouble translating the “for everyone” part as “tukuy runapaq” and software is an English loan word in Spanish, so it might as well stay a foreign loan word in Quechua. But we seem to get stuck on the word “free”. Apolinario doesn’t like the word “qespisqa” which is what the dictionary says is “libre” in Spanish. It comes from the word qespi which means “clear” or “glass”, but also means “to liberate”. Shara seems to thing that “qespisqa” is appropriate, since it is used in the quechua version of the Peruvian national anthem to describe the “liberation” of the people. In the local dialect, quespisqa means “to raise up” (subir) and can be used to describe building a house, but in the Cuzco dialect the word is closer to the word “free” (libre). Daniel as a free software advocate is adamant that we need a word for “free software” and it needs to be in the title of our manifesto. But Daniel doesn’t speak Quechua, so he is coming to the project as a free software advocate (or “linuxero” as they say here in Spanish). I won’t go into all the political and philosophical reasons why it is important that the project be under the banner of “software libre” or “free software”, but suffice it to say that a number of people have no interest in supporting the idea of Quechua software if it just gets called “software for everyone”. Microsoft can call their version of MS-Office in Quechua the same thing. The important thing is that we are advocating for a fundamental transformation of how people approach technology and intellectual “property” in general. I agree with Daniel that we need to use the word “free” but I don’t want to step on anyone’s toes.

Sadly we haven’t decided on a URI for our project. I would be perfectly happy to keep using, but I think that it is important that the group decide on a new name. Quechuistas tend to not like English loan words like “ciber”. In Bolivia, where everyone calls places to use the internet “ciber-cafes”, it seemed to capture the essence of the project, but in Perú where they are called “cabinas publicas”, a name like “ciber-runa” carries more of a foreign connotation. Some jerk in Russia has already registered the name probably in hopes of milking some money out of someone. Apolinario has suggested which means “in Quechua”. I would like a word which fuses Quechua with a word of technology. For instance:, short for “runasimi software”.

I find myself constantly trying to balance the fact that I have very strong ideas about the direction that I want this project to take, but it will never be a viable project if there isn’t a network of people here in Peru who take ownership and carry it forward as their own. Ella Baker-style organizing is really hard work and tough on the ego. I’m not sure that it is in my disposition to work by consensus or to relegate myself to the backseat in anything that I do. The problem is that I have the clearest sense of what needs to be done, so Daniel, Manual, Apolinario, and Shara have naturally deferred to me. I’m hoping that we can have a general meeting soon so our plans will get hammered out in a more collaborative manner.

I was all set to go to Cuzco today with Daniel to try and solicit some Cuzceño collaboration. Then Daniel discovered that he had too much work to do to be able to leave, so Daniel arranged for me to hitch a ride in the car with some of his friends. Unfortunately, his friends decided to start drinking in celebration of the Fiestas Patrias. Our one oclock departure time turned into 5 oclock, at which point I decided that there was no point in leaving because everyone that I wanted to talk to Cuzco was probably getting just as drunk, or out of town, or simply in no mood to talk about free software projects in Quechua while in the throes of holiday-making.

Today I spent a frustrating day watching Peruvians marching down the main drag in their respective corporate identities. I have earnestly studied the social function of corporativist parading in the plaza, but I still find the act extremely distasteful when used in the context of nationalist celebrations. Each person is subsumed as a representative of the power and prestige of their institution or social grouping. It was fascinating to watch as each university professor at UNAMBA donned the same color of suit and attached the same Peruvian national flag button to their lapel in order to march down main street. If a university professor didn’t have the proper coal-black suit, they would loan him or her the proper attire. I ran into one of the English professors and she asked me what I thought of the Fiestas Patrias. The proper thing to do would have been to smile and exclaim my approval for the event in which they were all so enthusiastically participating. Instead, I launched into a feminist critique of the contest to crown a queen for the festival. Then in my usual historian mode, I started analyzing the social function of the festivities and relating it to other nation-making projects of the past which had served to marginalize certain groups under the guise of “progress”, “modernity”, liberalism, the egalitarianism of citizenship or whatever Foucaldian discourse is used in the service of nationalism. In typical Latin fashion, she sweetly smiled, then asked me if I wanted to march with the other English professors to help them promote their department. I had to laugh. Afterwards I related the incident to Daniel and he thought it was just as funny as I did, so at least someone else grasped the irony.

So that is how the project stands right now. All in all I think we are making good progress although we have done very little actual translation so far. I’m a little apprehensive because everyone who has expressed interest in working on the translation is too busy to dedicate much time to it. I’m also a little concerned because I seem to be developing a dry cough which won’t go away. I’m going to have to see a doctor, but more pressing is the need to wash my clothes. I keep putting it off, but today it looks like I’m going to spend the morning scrubbing the month-of-wear smell out of my clothes. That means I’ve got to go and find something other than dental floss to use as a clothesline.

Take care,

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s