Friday, October 18, 2019

Science as Religion


What is science?


Empiricism


The word science actually means knowledge and refers to our understanding of our world, our universe. A closely related word is ontology which more literally means the study or understanding or logic of what is - but this has a closer relationship to what an individual understands about their world, including the internal and external aspects of our physical environment as we experience it, including the linguistic, social, cultural and ethical aspects.

The scientific method is often described as being an experimental method, or more formally empiricism. But there is also a flipside to this which is theory.


So how does this work?

Basically we are trying to understand the world, whether we are a scientist doing empirical science or an infant learning language and ontology. This means we collect data (who, what, when, where) and seek to develop explanations (why and how). These are all good questions to ask, and it is the last two in particular that are what scientists and infants ask incessantly.

The explanations (theories) usually involve assumptions (hypotheses) about things we don't directly know, rules (laws) that we think connect different parts of the data (observations),  reasoning (logic) from the hypotheses to show that the known data is consistent (verification),  and then exploring consequences of the theory (predictions) that take us into areas we haven't yet observed - and finally we run experiments to confirm or disconfirm these predictions (empiricism), and the theories may typically allow us to build new constructs (apparatus and technology). 

Figuring out how to apply these theories and adapt these technologies to our needs is called applied science, while design and engineering take over the eventual building and ongoing manufacturing of the resulting products.

But when people believe their theories, that is a matter of faith not science, and many social models and theories are more like religions in terms of being believed with religious fervour and defining a prescriptive way of life. Unfortunately belief in Evolution also has more of the characteristics of religious belief than scientific empiricism - and that does more harm than good to the scientific case for Evolutionary Biology.

Popper vs Kuhn


This scientific method goes back millenia, to people like Leonardo da Vinci, Copernicus, Galileo, Kepler, Newton and so on. Some times new formalisms, new forms of mathematics are developed as part of this process - like the development of calculus by Newton. This leads also to the idea of pure and applied mathematics too.

The empirical method was very clearly formulated by Karl Popper, who characterized real science as developing theories, making predictions and testing them experimentally. He quite sharply delineated proper science as being capable of invalidation, and proper scientists as actively making predictions into the unknown that could potentially invalidate their theories.

It is anathema to actually try to bolster and maintain your theories without making such predictions and experiments, or in the face of contradictory evidence from the experiments.

Kuhn took a more practical approach - scientists in practice are human and tend to defend their ivory towers, patching theories while controlling who gets employed and who gets published to keep down outsiders with competing theories.  Basically abandoning a theoretical approach and adopting a new theory comes in one of two ways: a paradigm shift where the results and contradictions are too big to ignore; and dying out literally as the old vanguard passes away.

So how do we compare to theories. Again there are basically two ideas that go back millennia: the first relates directly to the empirical method - we take the two theories to points where they make different predictions, and perform experiments to see which (if any) predictions are borne out; or if we can't make such predictions, or the old theory has been patched to give the right results, then we use the parsimony criterion that the simples theory is best (Occam's Razor).

Interestingly, supervised Machine Learning works in a similar way, making changes in a model anytime it gets things wrong, until it is right on all the data - but in the end we could just remember all the data to be always right (instance/example based learning). But this has a danger of overfitting, and the simpler model is to be preferred - and we must also make sure we always test on unseen data to make sure we are not just overfitting to the noise, and eventually we must take the model to different places and different sources of data to make sure we aren't overfitting to artefacts of the way (who, what, when, where, why, how) we collected the data.

Chomsky: nature vs nurture


One good illustration of this is the question of whether language is learned or innate. This goes back decades, and in particular Noam Chomsky believed that language was inherently innate, with specific organs for things like syntax and phonology that were "as real as the heart or the liver".  But the experiments were not performed to verify this, nor were the biological predictions formalized.  Chomsky's Principles and Parameters theory was humbly known as standard theory in the 1970s and 1980s (with Chomsky regarded as the founder of standard linguistics by his followers). It was based on knowledge and patterns from all known languages, and could generally be massaged to fit the inconvenient data that sometimes emerged.

The contrary view is most clearly represented by Jean Piaget (known as the father of psycholinguistics) who studied how studied all aspects of how the child learned about his world, language and culture (publishing over 20 books on the experiments exploring different facets of his constructivist theory). Other important names on this side of the ledger include George Lakoff  (known as the father of cognitive linguistics) who emphasized the role of metaphor and analogy in the way language was learned and evolved. There is an excellent volume published in 1979 about the Debate between Chomsky and Piaget, and an excellent volume called Metaphors we live by published by Lakoff and Johnson in 1980.

Of course one way to explore this question empirically is to look at programming computers to learn the way a human baby does. David Powers and Chris Turk published Machine Learning of Natural Language in 1989 based on a decade of experimentation of this kind by Powers, building on Kenneth Pike's theories of phonology, tagmemic grammar and universals of human behaviour, as well as hypothesizing separate recognition and production grammars, grounding the language learner in a simulated robot world, and predicting things like mirror neurons – connecting this to Turk's idea of anticipated correction and providing a neurologically plausible computational basis for the emergent constructivist cognitive linguistics approach. 

In the 1980s, Cognitive Science emerged as an interdisciplinary group of disciplines as people realized that is was important to cross the disciplinary boundaries between linguistics, psychology and neuroscience, to connect to the theoretical explorations of philosophers, and to exploit the computational modelling capabilities of computers - that actually allowed cognitive and evolutionary theories to be tested in ways not hitherto possible. Moreover Cognitive Linguistics emerged during the course of the decade as a direct rebellion against linguistic theories that didn't believe in learning, with Lakoff's 1987 book Women, Fire and Dangerous Things being an important landmark in defining the emergent field.

In the 1990s, Luc Steels took this language learning approach in a new direction, looking at the way language is evolved by cooperating robots working in a physical environment. This led to exciting collaborative experiments in the area of robotic language games, and a whole new perspective on the nature of language.

In the 1990s, David Powers broadened out from a focus on syntax, semantics and ontology into exploring the emergence of phonemes and morphemes in text and speech, leading to a self-organized multimodal hierarchy of linguistic processing that automatically learned to parse from phonetic features all the way up to phrases and clauses.

In the 1990s Noam Chomsky abandoned his 'standard' P&P approach, publishing his new minimalist program around 1992. However massive divides persist in linguistics to this day, and computational linguistics has moved out of linguistics into computer science - losing and ignoring the traditional psychological and linguistic insights.

In the 21st century, academic scientists and university department have unfortunately gone back into their disciplinary boxes and ivory towers, while research in artificial intelligence has largely been driven by commercial considerations. The rapid expansion of information technology and black box approaches to  neural networks has left little room for students to be trained in the relevant cognitive science, or to learn the language and ethos of the other relevant disciplines.

Dawkins: Evolution vs God


The second illustration we will use is the treatment of evolution in science, including both formal scientific writing and the popular writing of esteemed scientists.

In the penultimate paragraph of the introductory chapter to his 2019 book "Outgrowing God",
Richard Dawkins writes "all I need to say at present is that evolution is a definite fact: we are cousins of chimpanzees, slightly more distant cousins of monkeys, very much more distant cousins of fish and so on." Is this what we would expect from a top scientist?

In his previous writings Dawkins refers to biblical stories and characters as myths, while other people from similar periods are treated as historical. In "Outgrowing God" he does seek to justfiy this view by referring to literary criticism of the historical documents, and hypotheses of assertions. In the end, he admits the possibility that Jesus actually existed (and notes that the gospels were written long after Jesus death, and that people "wrongly believe" they were written by the authors they are attributed too - ignoring the internal evidence of Luke and John in particular, that there were stories circulating from early times, and identifying themselves in their texts).

There is considerable support for believing that key biblical characters existed. To start with the most obvious and superficial: Christians incorporate the name of Christ, and Semites (semitic races) incorporate the name of Shem, reflecting traditions going back to the named figures, with two large racial groups tracing their roots to Abraham (through Ishmael and Isaac). Other key biblical figures interacted with key political figures of the time - and generally the bible's view of Jews and Christians is not always positive: it tends to be no holds barred in a way that speaks to their existence. In several cases there is more evidence than for other historical figures of the same era. Thus it is prejudicial to call these mythical and others historical.

There are no testable predictions and scientific experiments we can perform here, but nonetheless there is in all cases evidence we can weigh.

On the other hand, acknowledging these people actually existed doesn't of itself mean that God exists, created the universe, or is interested in mankind, or has sent a saviour – although it does suggest that there are some purportedly miraculous events in the historical records that otherwise need to be explained away.

The essential point here is that evolution is also a theory where it is difficult to make testable predictions. It should not be held as a matter of faith - to do so is religion not science.

 Furthermore the theories of how particular animals might have evolved depends on myths called "just so stories" that take their name from Rudyard Kipling's (1902) Just So Stories for young children, stories like  How The Camel Got His Hump; How The Leopard Got His Spots; How The First Letter Was Written; How The Alphabet Was Made. These evolutionary "just so stories" are true myths, as nobody seriously believes that they represent actual historical events – although their authors hope there might be some similarity to what actually happened (but in some cases they seem more hysterical than historical).

One additional complication with dealing with evolution as a theory is that there is no well-defined theory, and indeed different people at different times mean different things by evolution.

Evolution vs Genetics


 Charles Darwin knew all about selective breeding, that is artificial selection, and that is where the name came for his theory of natural selection. His contemporary, Gregor Mendel formally studied artificial selection and effectively predicted the discovery of genes and the idea of genetic crossover of the parents genes – however, Darwin may not have been aware of Mendel's contemporaneous work. It seems he did have the journal volume in which the paper appeared, but the pages in that copy had not been cut (although that doesn't necessarily mean he didn't hear about it on the scientific grapevine or have someone else show him their copy). 

Work and publication of Origin of the Species did make Darwin question his Christian faith, or at least orthodox versions of it, as well as gaining him the criticism of some elements of the church –despite withholding his discussion of mankind as "so surrounded with prejudices" (letter to Alfred Wallace). He, however, didn't use the word "evolution" till Descent of Man, which did address this application to mankind and human races (and indeed the one use of any version of the verb "evolve" came right at the end of the extant editions of Origin of the Species).

Natural selection as a building block of evolutionary theory is well established, and these days not particularly controversial – having the character of making predictions that have been confirmed.  Of course, there is an artificial component to all experiments, but rather than direct breeding it has been possible to explore the changes in certain species engendered by changing environmental conditions or translating them to a different environment and/or ecosystem.

Creation Accounts and Myths


It should be emphasized that many Christians (and people of other religions) have no particular problem with proposed artificial and natural selection mechanisms, and the genetic processes that lead to evolution of species to better fit their changing ecological niches. Although transmutation of species, or evolution of new species, has proven hard to demonstrate (in the sense that new phenotypes and genotypes can still interbreed).

 But what of the the "myths" of the Bible, as Dawkins calls them: the "creation myths" and "genealogical myths". 

In many ways the Bible's "genealogical myths" are not much different from the "biological classifications" of Linnaeus.  The Linnaean taxonomy was a very important contribution, and the classifications did represent specific predictions about relatedness that have been tested (and much revised) based on genetic evidence. The family genealogies of the Bible did of course start off as oral history, and have been maintained more recently in written records.  Cluster Analysis and Principle Component Analysis of genome diversity across races is remarkably consistent with the biblical genealogies dating back to Noah, and it is also interesting to compare that with taxonomic/cluster analyses and geographical dispersion of linguistic diversity.

Genesis 1 and Genesis 2 actually provide two different accounts of creation - the first dealing with creation in its broader sense, culminating in the creation and dominion of mankind, while chapter 2 gives the story of Adam and Eve and lays the foundation for the fall into sin of Genesis 3 The first chapter has a somewhat poetic character and the seven days can be interpreted as eras (like the day of the dinosaur) and this is reflected elsewhere in the Bible (according to the book of Hebrews we are still in God's seventh day of rest). Similarly there is some evidence of editing of the lists of generations and choosing key figures to make a point. Matthew's grouping into three groups of fourteen generations from Abraham to David to Exile to Jesus is six of seven "weeks".  We are still in the seventh "day" of creation, the day of rest, according to the book of Hebrews, and we are evidently in the seventh set of seven "generations" according to Matthew. In fact, Genesis 2:4-5 uses the same word "day" to encapsulate the entire period of creation up to the point where man was created.

The Genesis 1 account thus doesn't purport to be day by day history, but reading through it sounds remarkably like a bystander viewpoint on different stages of creation. Many Jews, Moslems and Christians will tend to believe that it is not intended to be taken literally - and this view was emerging around the time that Darwin published Origin of the Species and accounts for the less than expected criticism across the church - in a sense it was overshadowed by the more general controversy over literary criticism of the biblical texts.

The Genesis 2 account reads very much like a just-so-story or fable aimed at teaching a particular lesson, although in the bible many such stories are painted as literally happening, sometimes there is debate as to they are meant to be interpreted literally or as parables. But in a sense this is irrelevant, as an omnipotent God could choose to teach Adam his place in the world, and that of his wife/woman (no separate words in Hebrew or Greek), by literally enacting this Adam's Rib story. 

The Genesis 7 account of Noah and the flood is also open to interpretation. In particular what is meant by "world" or "earth".

Here it is important to remember that when it was composed, the most it could be expected to represent was the "known inhabitd world" centring around the meeting point of Africa, Europe and Asia (the Middle East): the "new world" had yet to be discovered, and even that was just the Americas - the southern continents of Australia and Antartica are not part of either the old or the new world! Furthermore, the word "earth" (the expression in this chapter is "face of the earth" or "face of the ground") and can refer to the dirt and soil that is the foundation for life. There is no claim or words that imply "global flood", and a global flood wouldn't have been needed to wipe out mankind when still confined to the region of the Africa–Middle-East conjunction (actually part of Africa until relatively recently).

Similar considerations apply to the Genesis 11 account of the Tower of Babel and the emergence of different languages. Interestingly, God decides to "confuse their language" and does this by "scattering them abroad from there over the face of all the earth" (diaspora).

Initially, that need not have included crossing oceans to other continents, and humans clearly did migrate across land bridges and cross rivers and straits, and 60-70,000 years ago evidently accidentally or intentionally reached Australia without a need to cross more than 100km of water – and indeed the evidence suggests that the earliest Australian aboriginals were also able to cross to Tasmania, which was also accessible via a land bridge even as little as 30,000 years ago. Indeed these Tasmanian aboriginals seem to be genetically divergent from those that arrived in the last 10,000 years and one way of another the earlier race disappeared from the mainland. Even today, crossing from Siberia to Alaska requires bridging only around 40km of the Bering Strait, and there is evidence of accidental crossings of animals on ice floes.

Evolution of the Gene - the Science and the Fiction


In general, evolutionary theory is itself highly dependent on myths and just-so-stories to illustrate how things might have happened, although without the miraculous element of the Adam's Rib account. Furthermore there is a problem with evolutionary being "saltatory", that is jumping faster than evolutionists can easily account for - and the flipside of this is "missing links". But this doesn't mean evolution of the species or of mankind in such an undirected way is wrong, just that it is unproven - although it does beg further assumptions: e.g. the assumption of some kind of big-bang to primordial soup starting point (or similar). Of course their are other possible assumptions, e.g. the assumption of a God that directs the course of speciation – God could potentially directly reuse (genetic) components of earlier species and/or could control environmental conditions to more indirectly direct the course of evolution.

Many Christians accept this kind of post-genetic evolution in some form, but with God playing a role at points in the story: theistic evolution. The case for atheistic evolution is, however, not helped by its frequent personification as Evolution with a capital-E in the role of intelligent agent.

Knowledge of breeding and artificial selection techniques, and indeed theories that predicted some sort of genetic basis for speciation, were already extant in Darwin's time – and the predictions often did not tend to go beyond what was already known to be possible, and as noted above involved direct or indirect manipulation of either the breeding or the ecosystem.

Furthermore the big question that is somewhat less in focus, but nonetheless fundamental to big-E Evolution, is how the gene evolved, as well as related questions such as how and why sexual dimorphism evolved.

And taking a step back from that, there is the question of how  DNA, RNA, proteins and the like emerged from the mythical primordial slime.  Some relatively simple organic molecules (components of DNA, RNA and proteins) have emerged in test tube experiments, but we are a long way from explaining this.

From a science fiction perspective, if there are environmental conditions and pressures that could evolve such complex molecules, does this entail that compatible molecules will evolve in different environments (different planets in different galaxies).  SF theories of parallel evolution explicitly assume so - and it is not an unreasonable expectation of this kind of pre-genetic evolution.

But interbreeding between species on Earth is not possible (by definition of species), so interbreeding between earth and interplanetary or intergalactic species is also very unlikely without either some prior contact (genetic exchange/panspermia) or some kind of deliberate engineering (genetic manipulation/splicing) or some kind of common creator (intelligent design/theistic evolution).

My Books

So it is not likely that "parallel evolution" could lead to compatible species on other worlds.  Quite apart from the general question of "life" on other worlds, the inability of species to interbreed even on earth makes that difficult without some sort of genetic manipulation. But this is not the only questions - can we even expect to eat the plants and animals on another earth-like planet? Maybe! 

 The interesting question of compatibility at the level of DNA and RNA is one which is much more feasible as they are the only molecular building blocks we know of that can before their various genetic and messenger functions - my Paradisi Chronicles Casindra Lost series finds an affirmative answer to this on New Eden!


My Paradisi Lost stories

My Casindra Lost stories feature genetic engineering and an emergent AI 'Al' and a captain who is reluctantly crewed with him on a rather long journey to another galaxy - just the two of them, and some cats... There's another one, 'Alice' that emerges more gradually in the Moraturi arc. The Paradisi colonization aims to preserve the pristine ecosystems of New Eden, restrict mining to the other planets and asteroids of the system, and genetically modify people to suit the ecosystem rather than overwhelm it with introduced species: https://paradisichronicles.wordpress.com/

Casindra Lost
Kindle ebook (mobi) edition ASIN: B07ZB3VCW9 — tiny.cc/AmazonCL
Kindle paperback edition ISBN-13: 978-1696380911 justified Iowan OS
Kindle enlarged print edn ISBN-13: 978-1708810108 justified Times NR 16
Kindle large print edition ISBN-13: 978-1708299453 ragged Trebuchet 18

Moraturi Lost
Kindle ebook (mobi) edition ASIN: B0834Z8PP8 – tiny.cc/AmazonML
Kindle paperback edition ISBN-13: 978-1679850080 justified Iowan OS 

Moraturi Ring
Kindle ebook (mobi) edition ASIN: B087PJY7G3 – tiny.cc/AmazonMR
Kindle paperback edition ISBN-13: 979-8640426106 justified Iowan OS 

Author/Series pages and Awards

Posted by at
Labels: , , , , , , , ,

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)