Chapter 0: The Power of Babelfish


(S. Joshua Swamidass) #1

Faith Across the Multiverse by @AndyWalsh, 2018 by Hendrickson Publishers, Peabody, Massachusetts. Used by permission. All rights reserved.

I have a particular affinity for Babel fish [translator fish from The Hitchhiker’s Guide to the Galaxy] because I’ve spent much of my career filling a similar role within the sciences. I wound up learning a lot of math for a graduate student in microbiology, so I spent a lot of time teaching statistics to my fellow biologists. That helped me get a postdoctoral fellowship facilitating collaboration between computer scientists and biologists by pointing out where their ideas overlapped and, just as importantly, where they thought they were talking about the same thing because they were using the same word, but they actually had very different concepts in mind. And now I work for a public health software company, translating between public health users and software engineers and teaching everyone a little math and basic biology along the way.

My favorite word from those cross-discipline conversations is “vector.” To someone with a math or physics background, a vector is a quantity associated with a direction, such as wind velocity. To a computer scientist, a vector is a collection of data elements that may or may not be numeric. To a molecular biologist, a vector is a circular DNA molecule used to add external gene functionality to a cell. And to an infectious disease specialist, a vector is an animal that carries a disease, like a mosquito or a rat. Many folks are aware of the different, domain-specific meanings, but even experienced interdisciplinary researchers can get caught thinking about numbers while their colleagues mean ticks. And so having a “multilingual” Babel-human like me around helps keep conversations on track.

Language barriers are also a common problem in science fiction. Aliens should speak a variety of languages. Making every character a polyglot is unrealistic, and constant fumbling over language barriers would get in the way of good storytelling. Conjuring Babel fish or Universal Translators or literal magical spells simultaneously acknowledges and shelves the issue.

Translation is not science fiction. Making it instantaneous and perfectly accurate is the unrealistic part; sometimes we have to wait for the translation. Translating is not just a matter of swapping one word at a time for its equivalent, which is how something like the Babel fish apparently works. Anyone who has studied a foreign language has encountered idiomatic phrases whose meaning is not captured by a word-for-word translation. For example, I made my German teacher giggle while practicing pet-related vocabulary because my grammatically correct and word-by-word accurate Ich habe einen Vogel (“I have a bird”) was simultaneously an admission to having bats in my belfry, so to speak.

Unlike most science fiction, the film Arrival digs deep into matters of translation. Aliens arrive on Earth for the first time and linguist Louise Banks leads a team to establish communication with them. Learning individual vocabulary words and idioms is challenging enough without a dictionary, but Louise recognizes the possibility of even deeper problems. The aliens’ experience may be so different from ours that they think about different concepts. Do they have a notion of war? Do they distinguish between tools and weapons? These are the immediate concerns of world leaders wondering if these aliens come in peace, but Louise eventually discovers that the conceptual gulf runs deeper still.

Even human languages don’t overlap fully in terms of the concepts they can represent. If we give it any reflection at all, we probably think of our languages as complete. Sure, maybe we need to invent new words when we invent new technologies, like the telephone or Facebook. But for regular ideas, surely we must have the words to say what needs to be said. Only, how would we talk about the things our language lacks the vocabulary to describe?

The Germans have a very useful word: Weltschmerz. Translating the parts of this compound noun into English yields “world pain” but a more faithful translation might be the feeling one experiences upon recognizing the divergence between reality and an ideal vision of the world. English lacks an equivalent word; the closest match might be Charlie Brown’s exasperated “Good grief!” But even that is more of a groan, signifying Charlie Brown is experiencing Weltschmerz without actually naming it. All languages differ in which concepts they can readily express with a single word or common idiom; even fundamental features like the number and kind of verb tenses can vary. What is easy or hard to express in a given language influences how speakers of that language talk and possibly think.

Another example that might be more familiar to Bible readers is love. We can say a lot about love in English; poems, songs, and tales of love abound. Love is such a fundamental part of the human experience that we might think it would be a foundational part of any human language. In fact, the ancient Greeks had several words to differentiate experiences we lump together as love. These include the familial bond between siblings, parents, and relatives; the brotherly affection shared by comrades-in-arms; and the romantic or physical connection that we generally mean when we say one is “in love.”

Most of the time English speakers don’t consciously experience a deficiency or limitation in their language regarding love. We blithely say “I love you, my dear” and “I love you, dad” and “I love you, man” and “I love you, delicious chimichanga” and generally everyone knows what we mean. But sometimes we don’t say “I love you” when maybe we should. Perhaps we sense our feelings for our spouse are not the same as our feelings for our children but we lack the language tools to express that nuance succinctly.

In Arrival, Louise Banks addresses this conceptual gap between her language and the aliens’ by combining written language with demonstrations. She and her team act out the words and sentences as they speak and write them. In this way, they build shared experiences with the aliens so that their communication has something to reference. Without that experience, the two parties might wind up using the same words but internally connecting them to very different concepts. As an example, she brings up the Sanskrit word गविष्टि, which some linguists translate as “an argument” while she prefers “a desire for more cows.”

Do not forget about the The Babel Fish Argument for God.


(Dan Eastwood) #2

Don’t you mean the Babelfish argument for the non-existence of God? :wink:


(S. Joshua Swamidass) #3

The multilingualism of science is surprising and important. It takes time to get used to in interdisciplinary work like computational biology.

@AndyWalsh you like to use “vector” as an example of this. Can you explain?

Do any of the scientists (or other scholars) have examples of words in their field used differently in other field or in common usage? @art @John_Harshman @Philosurfer @deuteroKJ @evograd @nwrickert @T_aquaticus @PdotdQ @pevaquark


(Neil Rickert) #4

Mathematics has many such terms.

Perhaps the worst example is “normal”.

A normal subgroup S is a subgroup such that aS = Sa for every a in the group.

A normal topological space is one where any two disjoint closed sets have disjoint open neighborhoods.

A normal line in geometry, is a perpendicular line.

Then there’s the normal probability distribution, sometimes known as The Bell Curve.

I guess we have too many technical terms in mathematics, and we run out of ideas for naming them. So we reuse names.


(S. Joshua Swamidass) #5

How does code mixing and switching work in mathematics? And other fields?


(Neil Rickert) #6

It usually goes quite smoothly. We don’t even notice it. Human language abilities seem well tuned for using context as part of how we recognize meaning.


(Kenneth Turner) #7

cult, myth, literal, apocalyptic, prophetic, gods, image of God, demons, messiah, etc.


(S. Joshua Swamidass) #8

Can you elaborate some on that? What are the differences in definitions and usage?


(Kenneth Turner) #9

Cult - official religious system (e.g., Leviticus describes Israel’s cult), not a heterodox group

Myth - there are several technical definitions (e.g., involvement of the supernatural realm in human and earthly affairs), most of which do not assume falsehood or non-reality

Literal - often more about authorial intent (literary meaning) than literalistic sense

Apocalyptic - a genre involving strange images and speaking to hidden divine activity overseeing history, with a view to a decisive end to the present state (but not just about that like we see in a sci-fi sense)

Prophetic - not primarily about simple prediction/fulfillment, but a call to covenant fidelity; even the predictions/fullfillments are rarely simple

gods - the Hebrew 'elohim is used for beings that don’t all possess the set of attributes normally thought of when we see g-o-d(-s)

image of God - I’m thinking of the debate between ontological, relational, and functional/vocational definitions. Most in my subdiscipline (biblical theology) tend toward the functional/vocational sense; most systematic theologians and philosophers tend toward the ontological sense (e.g., a specific attribute shared by God and humans).

Angels/Demons - the ANE/OT taxonomy of supernatural beings is much more complex than the simple angels & demons usually thought about

Messiah - a specific term related to the Davidic covenant, and can have more than one fulfillment/anti-type


(John Harshman) #10

If I recall, physicists (?) use “drift” in a completely opposite sense from ours.


#11

I have experienced linguistic disconnect when talking to the lay audience using the words theory and error.

Something like the theory of gravity does not mean that gravity is “just a theory”.

Simillary, my experiments have errors in them does not mean that the experiment is wrong.


(Andy Walsh) #12

In my biology studies, I’ve worked with both ‘vector’ organisms like mosquitoes, ticks and mice that transmit disease, and ‘vector’ plasmids that transmit genes. In those contexts, the meaning is basically the same when used as an adjective. But it is also common in both contexts to just use ‘vector’ as a noun without qualification.

In physics classes, a ‘vector’ is a quantity with a direction, and at least in elementary classes that direction refers to the three dimensions of physical space. Mathematically, you can represent a vector as a list of coordinates, which leads to an abstraction of a vector as a list of ‘coordinates’ in arbitrary spaces that may not be physical. Thus in computer programming a ‘vector’ might be a list of text strings without any obvious reference to a concept of space.

As @nwrickert pointed out, we can often handle these context-dependent meanings just fine. In fact, we often don’t notice we are doing it. So when someone calls our attention to it with a pun or double entendre or deliberate misinterpretation, we can be surprised and amused. “The other day I shot an elephant in my pajamas. How he got in my pajamas I’ll never know.” And so on.

At the same time, when I was a postdoc working with computer scientists and biologists who were exploring computational biology for the first time, I did occasionally have to translate overloaded technical terms like ‘vector’ or at least remind folks there were multiple meanings.

The other interesting bit of interdisciplinary translation was at a different level than individual words. It also became clear that different disciplines have different ideas of what it means to have publishable results. In computer science, the emphasis is on new methods and so it is common to publish about a new technique as applied to well-known data sets. In biology, methodology papers are rarer and it is more common and more acceptable to publish the results of applying a well known technique to a new data set/organism/cell type/context.


(Daniel Ang) #13

Exactly. The confusion over the word “theory” is amusing to watch. Sometimes, in an effort to correct the common misconception that theory = untested speculation, scientists go to the other extreme and state that theory = verified scientific hypothesis with lots of evidence to support it. Whereas in my experience, in reality scientists use the word theory pretty loosely, with little regard to its evidential status - it seems that the unifying feature of all theories is “scientific statements containing many mathematical symbols” (as opposed to simple experimental statements).

This also illustrates how most people don’t understand that all science is an approximation with errors at some level. Scientists often work with models with varying levels of accuracy, precision and usefulness depending on the regime they are tested in. Which is why scientists don’t always use the word “wrong”, but rather “inapplicable” or “imprecise” - which is different from “bad science”.


(S. Joshua Swamidass) #15

@AndyWalsh, you framed your book around four different languages in science: math, physics, biology, and computer science. Of course, each of these fields has subfields with their own languages too.

This was an instructive and coherent way of framing your book. Just the framing itself is a great way to introduce students to the diversity of science (@Faithdefender, are you reading along?).

What brought you to this idea?

You cover several theological themes through these


(Daniel Ang) #16

There is more to this than just needing to use foreign words to describe certain concepts. It reminds me of the Sapir-Whorf hypothesis: that language can affect our perception of the world.

Rather than delving into technical linguistic details, as someone who speaks both Indonesian and English fluently, I can identify with this at a personal level. There are many words in Indonesian which simply cannot be translated into English, even if you try to use multiple words to describe its meaning. It’s not just about overlapping concepts - some Indonesian words can instantly evoke in my mind a host of very Indonesian cultural associations. This also explains why Indonesian humor is very different from English language jokes. Sometimes even just a comic character using certain words to express a simple idea in Indonesian can evoke laughter.

Another interesting facet of this is my experience with Singaporean English, or Singlish - a unique Singaporean pidgin of English, Malay, Hokkien and Tamil (the movie Crazy Rich Asians has some glimpses of this). In my 4 years living in Singapore, I never really became a “native” Singlish speaker, but I did adjust my previously Canadian accent to a rougher, Malay accent tinged with various Hokkien/Malay words whenever I spoke with my friends or ordered food. Even today, whenever I meet someone from Singapore in the US, I semi-consciously switch to this accent. It instantly opens up an unspoken world of cultural associations and ways of expressions that just isn’t there when I speak regular, American-accented English. I feel there usually is an instant jump in the level of intimacy and trust in the conversation.

It would be interesting to hear other people’s experiences with bilingualism.

Bilingualism and Scientific Language

There is also a parallel of this in terms of “scientific language”. I often try to explain what I do for my PhD to all sorts of people. Often I have to create very simple analogies or cartoons to explain why we care that the electron is round or not. But whenever I meet someone who’s a fellow atomic physicist, or had previous experience in it, I can instantly convey my explanation in a few sentences. Besides a shared understanding of some atomic physics terms and concepts, there’s also a number of common topics and “paradigm cases” that we would both know and regard as important. This sort of unspoken understanding is only conveyed after years of interacting with fellow physicists in the field. Basically, if you speak the language of a specific scientific community, you establish trust with people in that community.

(Interestingly, this can also be faked, to some level, as sociologist of science Harry Collins demonstrated when he embedded himself in the LIGO collaboration for many years. This raises interesting questions about perception of expertise and understanding. It also shows that speaking a scientific language may not be exactly the same as understanding and being able to apply the concepts themselves.)

Naturally, it seems to me that some of the heated debates between philosophers, scientists, and theologians (or ID vs. TE vs. YEC and so on) is partially exacerbated by the lack of a common language. We even see this happening in this forum - “show me the peer-reviewed papers!”. Perhaps one way to achieve peace between these camps is to see whether we can establish some common vocabulary when we talk about God, the Bible, faithfulness to Scripture, and so on.


(Steve Schaffner) #17

You’ve touched on one of my pet peeves. It’s not just random scientists who make that statement about “theory” – it appears in semi-official statements, and never based on any kind of linguistic study. From the National Academy of Sciences: “The formal scientific definition of theory is quite different from the everyday meaning of the word. It refers to a comprehensive explanation of some aspect of nature that is supported by a vast body of evidence.” That statement is just nonsense. There is no formal scientific definition of theory, and no requirement for evidential support. The Georgi-Glashow SU(5) model, for example, is a theory of grand unification, but it’s a theory that’s contradicted by evidence, not supported by it. And this leaves out the second, distinct scientific use of “theory” as a mass noun, in which it refers either to a set of mathematical machinery (e.g. quantum field theory, used in specific theories like quantum electrodynamics, or diffusion theory, used by Kimura in formulating the neutral theory of molecular evolution), or to the entire body of theoretical work in a field (solid state theory, evolutionary theory).


(S. Joshua Swamidass) #18

@AndyWalsh you also talk about statistical distributions in this chapter, and point to exponential distributions as an example of how something in science fundamentally clarified a key theological concept for you. Can you tell us more?

Does any one else remember this?


(Andy Walsh) #19

Yes! The Sapir-Whorf hypothesis (and its controversial status) was on my mind a lot when working on this chapter. I went back and forth a few times on the passage you quoted specifically, not wanting to come across as endorsing the strongest version of Sapir-Whorf since in my reading there does not seem to be a consensus behind it, but also wanting to help folks reflect on their own relationship to language. I have no idea if I struck the right balance, but it is exciting to me that it prompted you to share your experiences with Indonesian and English, so thanks for that. (And now I have another reason to catch Crazy Rich Asians.)

I’m reminded of this TED talk about the role of tense in different languages and how it might influence thinking about the future (and consequently saving money).

And you’re absolutely right that accents, word choice and other subtle features of language can signal so much about culture, shared experience, and community even between strangers. Even as a simple monolingual English speaker, I run across this when certain words and accents remind me of my childhood in New Jersey, or when constructions like “needs cleaned” remind me that my children will be western Pennsylvanians in ways I never will be despite living here for 13 years.

I want to pick up on your other point about language in science communities in another post that also ties in to @swamidass 's questions.


(Andy Walsh) #20

So, the answers to both questions are related, and touch on what @dga471 mentioned about how shared language and concepts can signal membership in the community of a scientific (sub)discipline.

In the book I tell the story of a Bible study where we were wrestling with the idea that Paul and other New Testament writers wrote as if they expected to Jesus to return at any moment, within their lifetime. And here we are 2,000 years later, faced with the paradox that we are supposed to be just as expectant, yet surely that 2,000 year wait should provide some information about how long we can expect to continue waiting.

It’s a question I had come across before, but this particular Bible study took place while I was taking a probability theory class. Just as crucially, one of the other study members was a mathematics PhD student. In that moment, it clicked in my head that an exponential probability distribution has the needed property of ‘forgetting’ that 2,000 year wait. And in an example of the kind of shorthand Daniel was talking about, all I had to say to my friend the mathematician was “It’s like an exponential distribution” and it clicked for him too. The whole thing was very satisfying, and it demonstrated for me the power of language as a tool. And the power of shared language to build connections, as my mathematician friend was pleased to find a biologist who could speak his language.

Now, oddly enough, even though that was essentially the inciting incident for the book, and even though the structure of a math section, a physics section, a biology section and a computer science section were present from the earliest days of trying to organize these thoughts into a book, the actual framing in terms of language came very late in the process. The book was always about translation, but for some reason it took me a while to actually be able to articulate that.

Also on these topics: the flip side of the community that builds around shared language and shared concepts is the idea of jargon and the use of language to exclude people who aren’t ‘in the know.’ Just about every editor will tell you that jargon is bad. And there is definitely some terminology in this book that could be called jargon–scientific jargon, theological jargon, even comic book jargon. But it was always my intention to include those terms and phrases along with explanations and definitions, in order to invite people into the associated communities.

I want to say to folks “Here in this group, we say ‘x’ when we want to talk about ‘y’. You probably already understand ‘y’ and maybe you call it ‘z’ in your circles; I think you will also enjoy being a part of the conversation about ‘y’ in this group and so I’m giving you the word ‘x’ to help you.” Maybe I don’t always succeed–forgetting which words will be unfamiliar, patronizing with a description everyone already knows–but that’s the overall intention.


(Larry Baxter) #21

Quite a brisk discussion here, and we’re only on the first chapter! As someone who did research for many years at the intersection of mathematics and biology, I had to smile at the example of vector with its many different meanings.

The examples you all cite really highlight for me the power for shared meaning to unite and clarity, and the lack or inconsistency of shared terms to form a divisive wedge. The number of arguments I see where the “combatants” don’t realize they are using the same word in totally different ways is staggering. Politically, the power to name an agenda or view has been seen to be of great importance in capturing hearts and minds.

Two examples came to mind for me in reading this chapter. The first was as a software developer. One of the books that gave me great insights as a programmer was one on Design Patterns. The author looked at use cases with very different approaches or applications but where the underlying concept was identical. By using the language of these patterns (singleton, class factory, publish-subscribe), one could convey an incredible amount of information in very few words. But use those terms with someone unfamiliar with the pattern and the eye will glaze over. The second example comes from within the church. Pastors and critics of churches alike constantly use terms like disciple, missional, leader, inerrant, reformed, without realizing the words have little meaning, or a very different meaning, in the ears of those they are talking to.

I remember years ago as a seeker/atheist hearing the gospel from a good friend who was an engineer. He said that the nature of Christ and his relationship to the Father was a mystery and apologized it would be nearly impossible to explain. I chuckled and replied that was a non-issue for me and rather easy to understand. He looked shocked of course and asked how that was. My reply was long the lines of God was obviously a multi-dimensional being far beyond our understanding, but Jesus was the projection of a higher dimensional God into a 3D+temporal space. The math of that instantly clicked between us. (Whether that analogy is a geeky expression of Modalism, I don’t know :wink: )