The American Time Use Survey recently released results for 2018. That makes 15 years of data. What's different? What's the same? Read More
Posted by sciencein
Nicholas Rougeux, who has a knack and the patience to recreate vintage works in a modern context, reproduced Elizabeth Twining’s Illustrations of the Natural Orders of Plants:
If someone told me when I was young that I would spend three months of my time tracing nineteenth century botanical illustrations and enjoy it, I would have scoffed, but that’s what I did to reproduce Elizabeth Twining’s Illustrations of the Natural Orders of Plants and I loved every minute.
The best part is that you can select flowers in the text or on the illustrations to focus on a specific parts, which makes descriptions easier to interpret.
Posted by blood (journal) retractionsin
[This is a guest-post by Justin Power, and the 3rd part of our miniseries on sign language manual alphabets]
In Guido’s most recent post in this miniseries on manual alphabet evolution in sign languages, he discussed the role of character mapping on networks in phylogenetic inference. He pointed out how we used this approach to infer evolutionary pathways of languages, and why this step in exploratory data analysis is important, given the complexity of the underlying signal in this data set.
In this new post, I take up the topic of hand-shape evolution in more detail, explaining some of the complexities involved in studying sign language evolution. I will specifically look at how we can identify both vertical and horizontal processes in the evolution of hand-shapes.
We know very little about how signs and hand-shapes actually evolve. There have been a few studies — most of them from decades ago — comparing American Sign Language in videos and dictionaries from the early 20th century with then contemporary forms (Frishberg 1975; Battison et al. 1975). One study in particular argued that, as a sign language emerges in a community of signers, crystallizing into a stable linguistic system, the signs evolve in a quasi-teleological way from earlier, more gesture- or pantomime-like forms to more language-like forms, cutting similar evolutionary pathways leading to more constraints on articulation and to general systematization.
But what happens (in this story) once sign languages become linguistic systems? Do they continue evolving, as happens in spoken languages? If yes, how? Investigating these kinds of questions was one of my motivations for tracking down historical examples of manual alphabets for over a dozen sign languages. The pay-off (besides the thrill of the treasure hunt) is that, by tracing hand-shapes through historical examples and comparing them with contemporary sign languages, we can infer the vertical and horizontal evolutionary processes affecting sign languages and hand-shape forms.
Vertical and horizontal aspects of hand-shape evolution
Consider part of the Neighbor-net from our paper (see Part 1) including the Austrian-origin and Russian groups in the figure below. Russian 1835 is the earliest manual alphabet in our sample published in Russia (St. Petersburg); and Danish 1808, in the Danish subgroup, was published in Copenhagen.
While the two manual alphabets are found in different neighborhoods in the graph, they share a number of hand-shapes, some of which were (and still are) shared widely throughout Europe, for reasons that we discuss in the main paper.
One such hand-shape represents the Latin / Cyrillic letter "A" in both Danish 1808 and Russian 1835, as illustrated in the timeline here.
Note the position of the thumbs at the bottom of the figure: in both early examples, the thumb is adjacent to the bent index finger. In an example from Danish SL in 1907 (and subsequently in 1926 and 1967), the position of the thumb has shifted across the index finger. For Russian SL, too, the position of the thumb in the contemporary hand-shape representing the Cyrillic letter A has crept across the index finger to the front of the fist (the hand-shape in the figure is my attempt to reproduce the source; see here for the real thing).
There are two points to note here in connection with evolutionary processes. First, these changes in thumb position appear to have a vertical aspect: as signers in a community used these hand-shapes and transmitted them to later generations, they also modified the forms in subtle ways, perhaps unconsciously in a process with analogies to sound change in spoken language.
Second, the changes also include a horizontal aspect: the forms evolved in similar ways, as the two signing communities converged on the same shape (apparently) independently, possibly due to similar articulatory or perceptual pressures. The horizontal aspect of this process contributes to signal incompatibility in the dataset underlying the network — the more convergence there is, then the less tree-like will be the Neighbor-net (in this case, the more spiderweb-like).
In addition to the preceding example, a typical case of convergence can be seen in the independent creation of similar hand-shapes to represent the Greek and Cyrillic letter "Г".
Beginning again with the main Neighbor-net in the figure immediately above, we see that Russian 1835 and contemporary Greek SL are found in different neighborhoods, with Greek in the French-origin group. The two languages, however, share the Г-representing hand-shape (the Russian form is from Fleri 1835, while the Greek form is, again, my own hand; see here for the real one). Because Greek SL is the only language in the French-origin group to share this hand-shape with the Russian group, there is a clear suggestion of a horizontal process that resulted in similar hand-shapes across unrelated languages. The most likely processes here are convergence due to the independent creation of iconic representations of the written letter; or lateral transfer — called borrowing in linguistics — via some historical instance of contact between signers of the two languages. [My intuition is for the former explanation.]
The final example deals with a clear case of borrowing. The figure below shows the time- / taxon-filtered Neighbor-net, including historical manual alphabets up to about 1840 (see Part 2), but only annotated with the relevant languages.
The two earliest manual alphabets in our dataset were published in Madrid in 1593 (de Yebra) and 1620 (Bonet). In neither case do we see any trace of a hand-shape representing the letter "W", which was not needed to represent these Latin alphabets. Later, too, manual alphabets published in Spain in 1815, 1845, and 1859 still did not include the letter "W". In contrast, in Austrian 1786 and French 1800 (as well as other languages), hand-shape forms representing the letter W are found in the earliest examples we have for those languages. Some 160–230 years later, however, we find similar forms for "W" in contemporary Austrian, French and Spanish SLs. We deduce that contemporary Spanish SL did not inherit the "W" hand-shape from the 19th century Spanish manual alphabets. Instead, the hand-shape may have been borrowed from some other language, possibly French SL given its influence on deaf education in Europe, or possibly later from the International Sign manual alphabet (also part of the French-origin Group).
As these examples show, there are different types of horizontal processes contributing to conflicting signal in the data set. Using the splits network graphs together with historical examples of manual alphabets, we can untangle the horizontal signal in many cases. The approach has also given us some insight into the evolutionary processes contributing to the diversity of contemporary sign languages, a topic that we plan to investigate more fully.
Cited literature, further reading and data
- Battison, Robin, Harry Markowicz, & James Woodward (1975) A good rule of thumb: Variable phonology in American Sign Language. In Ralph W. Fasold & Roger W. Shuy (eds.), Analyzing Variation in Language: Papers from the Second Colloquium on New Ways of Analyzing Variation, Part 3, pp. 291–302. Washington D.C.: Georgetown University Press.
- Bonet, Juan Pablo (1620). Reduction de las letras y arte para enseñar a ablar los mudos. Madrid: Francisco Abarca de Angulo.
- Fleri, Viktor I. (1835) Глухонемые, рассматриваемые в отношении к их состоянию и к способам образования, самым свойственнымих при. St. Petersburg:Типография А. Плюшара.
- Frishberg, Nancy 1975 Arbitrariness and iconicity: Historical change in American Sign Language. Language 51(3): 696–719.
- Yebra, Melchor de (1593) Libro llamado Refugium Infirmorum: Muy util y prouechoso para todo genero de gente : En el qual se contienen muchosauisos espirituales para socorro de los afligidos enfermos, y para ayudar à bien morir a los que estan en lo ultimo de su vida ; con un Alfabeto de S. Buenauentura para hablar por la mano. Madrid: Luys Sa[n]chez
Other posts in this miniseries
- Stacking networks based on sign language manual alphabets – introduction and principal networks used in our study
- Character cliques and networks: mapping haplotypes of manual alphabets – how we explored the principal signals in our matrix
This is the sixth in a series of guest posts by Arlin Stoltzfus on the role of mutation as a dispositional factor in evolution.
by Arlin Stoltzfus
The authors of TREE's hatchet piece imply that the theory of Yampolsky and Stoltzfus (2001) is somehow not new, citing ancient work from Dobzhansky and Haldane. In Box 1, they argue that this theory is part of "standard evolutionary theory," showing a 4-step derivation ending in Eqn IV, which is Eqn 2 of Yampolsky and Stoltzfus (2001), and informing the reader that this is based on "classical" results from Fisher, Haldane and Kimura, who are named, while Yampolsky and Stoltzfus are not named.
Yet, Fisher, Haldane, and Kimura did not make the argument in Box 1, did not follow the 4 steps, and did not derive Eqn IV!
Reactionary fringe meets mutation-biased adaptation
1. The empirical case
2. Some objections addressed
3. The causes and consequences of biases in the introduction process
4. What makes this new?
5. Beyond the "Synthesis" debate
Indeed, we discovered last time that Haldane and Fisher, who died before origin-fixation models emerged in 1969 (McCandlish and Stoltzfus, 2014), argued against the potential for mutation-induced evolutionary tendencies, and their argument was repeated for another 70 years.
What is the purpose of this deception? Why would the authors divert credit for new theoretical findings to dead people? Why would TREE give a platform to authors who violate norms of scholarly attribution and exhibit hostility toward a line of research they do not understand?
Welcome to the culture warThe context for this perverse treatment is the convoluted debate over the status of the neo-Darwinian "Modern Synthesis" of the mid-20th century.
This debate, which began in the 1980s, has heated up recently, as reformers have begun to organize, not just via the vapid Third Way, but more credibly via the coalition calling for an "Extended Evolutionary Synthesis" (EES). The EES Front receives funding from the Templeton foundation, which causes concern to many scientists, and sends Jerry Coyne into paroxysms of indignation. Traditionalists were inflamed in 2016 when the Royal Society hosted a reform-themed conference organized by reformers and outsiders.1 The tone of the traditionalists, their authority seriously challenged, has become strident and desperate.
It was in this context that the reactionary authors of TREE's hatchet piece identified my colleagues and I as part of the barbarian horde of "critics of standard evolutionary theory" ready to breach the citadel of Synthesis orthodoxy.
Why do I call this a culture war? Why not frame this as an empirically resolvable dispute between two theories, Standard Evolutionary Theory (SET) and the Extended Evolutionary Synthesis (EES)?
Let us consider this question. The reformers hold that the gene-centric view of SET
"fails to capture the full gamut of processes that direct evolution. Missing pieces include how physical development influences the generation of variation (developmental bias); how the environment directly shapes organismsâ€™ traits (plasticity); how organisms modify environments (niche construction); and how organisms transmit more than genes across generations (extra-genetic inheritance). For SET, these phenomena are just outcomes of evolution. For the EES, they are also causes."From Does evolutionary theory need a rethinkThat is, SET and EES are characterized with sets of "processes that direct evolution." To offer an "extended" version of SET is to extend its list.
This does not mean that SET and EES are testable theories: they might be flexible research programs or schools of thought that shift emphasis and welcome new ideas, so that each one could evolve into the other.
Of course, the distinctiveness of EES could be translated into falsifiable claims: if SET's list of causes is incomplete, as the EES Front claims, then an account of evolution based on SET would be empirically insufficient.
For instance, EES includes developmental biases among the "processes that direct evolution." By contrast, according to the dichotomy of roles essential to neo-Darwinism, variation merely supplies raw materials, whereas selection is the governing force providing initiative, creativity, and direction-- a distinction reflected in Mayr's statement (figure), or the quotation about the essence of neo-Darwinism from Gould (1977) that appears in the left side-bar of SandWalk (see also Gould 2002, pp. 137 to 146).
Thus, if some of the directionality of evolution is due to developmental bias, whereas the neo-Darwinian position is that all of it is due to selection, then SET fails.
Likewise, modifications to the environment via niche construction are quantifiable, and so are their feedback effects on evolution, e.g., there is some total non-living mass of constructed environments (anthills, beaver dams, coral reefs, etc) and this must have some quantifiable effects on evolution via increased biomass or species richness. A falsifiable EES theory could predict some evolutionary consequences of increased species richness due to niche construction, and if these consequences can be shown, SET fails.
Yet, we face a challenge with applying such a test: SET defenders have retreated from neo-Darwinism, and they do not agree to a falsifiable position excluding developmental biases or niche construction.
For instance, the 2014 EES-SET debate in Nature (Does evolutionary theory need a rethink?) never invokes neo-Darwinism, though Charles Darwin gets 14 mentions. The participants did not seem to disagree on scientific issues (see The Great Non-Debate on Evolutionary Theory). Wray, et al. (2014), defending SET, acknowledge the in-principle legitimacy of every ostensibly non-traditional EES mode of evolution, which they characterize as "new words, old concepts." They claim that SET is always being extended and revised, and that it already includes all valid ideas of EES, though they raise doubts about the importance of these ideas.
This flexibility is what makes the Synthesis so durable, as historian David Hull (2002) explains in the Oxford Encyclopedia of Evolution:
"Any criticism of the synthetic theory that turned out to have some substance was subsumed in a modified version of this theory. Instead of being a weakness, this ability to change is one of the chief strengths of the synthetic theory of evolution. As in the case of species, scientific theories evolve" Hull (2002) History of Evolutionary ThoughtWhat Hull is describing, of course, is not a genuine scientific theory, but something like a tradition or a school of thought-- the Synthesis-inspired Evolutionary Tradition (SET).
Provine, who wrote the seminal history of the theoretical foundations of the Modern Synthesis, said that it "came unraveled for me during the period since 1980," citing widespread neutral evolution, the abandoned concept of genetic homeostasis, the lack of a unified theory covering molecules and morphology, and the collapse of the "gene pool" view (see Provine's afterword to the 2001 re-printing of The Origins of Theoretical Population Genetics). The original Modern Synthesis held that the "gene pool" acts as a dynamic buffer of diversity that prevents any direct dependence of the rate of evolution on the rate of mutation-- a theory that no one accepts today (see The shift to mutationism is documented in our language). The original Modern Synthesis is a special case, not the foundation of contemporary thinking.
Indeed, traditionalists no longer defend the original theory, but defend a notional "Synthesis" partly by relying on cultural arguments (see Stoltzfus, 2017), e.g., Svensson (2018) argues that reciprocal causation (associated with niche construction) is part of SET because Lewontin, a person associated with the Synthesis tradition, wrote an essay about reciprocal causation, 2 decades after the architects of the Modern Synthesis declared victory without invoking reciprocal causation to account for any important features of evolution.
What about the EES? The architects of the EES understood a decade ago that the notional "Synthesis" was no longer a falsifiable theory, but more like a research program with "a hard core of theories immune to revision, surrounded by a protective belt of malleable theories." The hard core, they assumed, was selectionism and a reductionist focus on upward causation from genes and population genetics. So, they framed EES as an alternative to this bottom-up, gene-centric, selectionist approach, focusing on key causal concepts that emphasize organisms, development, non-genic inheritance and interactions.
That was a strategic mistake. The EES Front under-estimated the flexibility of SET and its apologists, who have already claimed all of the EES extensions. Wray, et al (2014) claim niche construction on the grounds that Darwin worked on earthworms, whereas Futuyma (2017) simply declares the idea to be obvious-- he always knew it but didn't bother to write it up--, which apparently means that Futuyma owns niche construction and can assign it to SET, given his special power to re-write the Synthesis every few years to keep it current.
The treatment of Fisher in TREE's hatchet piece again reveals SET as a flexible tradition, not a scientific theory. In his opening chapter, Fisher (1930) invokes the opposing-pressures argument, repeatedly arguing against any role for mutational tendencies in evolution, and concluding that the researcher who understands how genetics leads inevitably to neo-Darwinism
... will direct his inquiries confidently toward a study of the selective agencies at work throughout the life history of the group in their native habitats, rather than to speculations on the possible causes which influence their mutations.To accept mutation-biased adaptation is to revoke Fisher's theory.
However, traditions are more flexible than theories. SET defenders do not have to follow Fisher's position, but only to honor the forms of tradition, which they may do in various ways, e.g., by focusing on authority figures, or by simply re-telling the Synthesis story to accommodate new findings. For instance, the way to appropriate a theory of internal biases rejected by Fisher, Haldane, Wright, Simpson, Stebbins, Mayr, Ford, and so on, is to (1) discard the logic and purpose of the theory, breaking it down into parts, and (2) link some of those parts with tradition. This cultural activity cannot be distinguished from science if we have no standards for what constitutes a valid argument about scientific theories. For instance, if scientists do not have a clear understanding that scientific theories are not merely lists of parts, they will be susceptible to this kind of bad argument.
The other third wayBy way of disclosure, I am not a member of any reform movement. My primary criticism of the Modern Synthesis is the same as that of Nei, which is that the shifting-gene-frequencies theory gets population genetics wrong. The EES Front invited me to a workshop in 2017, which I enjoyed. In a concluding session, I argued-- extending informal comments and The Great Non-Debate on Evolutionary Theory-- that EES is not a cohesive theory but a list of reforms, and that confronting the notional Synthesis without changing the terms of debate inevitably turns the EES challenge into a cultural battle, a strategy that is not scientific, and that is sure to lose.
In response, one of the standard-bearers of the EES Front said something along the lines of "yes we are waging a culture war, and we will win!"
This comment split the room: half cheered, and the other half were stunned, an incident that prompted me to write Why we don't want another 'Synthesis'. The aims of science would not be served by setting up another intellectual monoculture so deeply attached to a totalizing modernist narrative that, merely to maintain the narrative, it will torture history, distort basic concepts, and sacrifice scientific accountability. When the authors of TREE's hatchet piece claimed evolutionary biases due to internal biases in variation, i.e., orthogenesis, on behalf of the Synthesis, this was merely the latest in a series of increasingly flagrant attempts to shift the goalposts to forestall valid criticisms of historically important ways of thinking.
The next sub-series of posts provides a basis for moving beyond the futile Synthesis debate, by taking a rigorous approach to history, maintaining a firm grasp of what theories mean and how to recognize them, and confronting rhetorical tactics that blur together scientific and cultural arguments.
5.1. Thinking about theories. Here we (1) distinguish theoryC (concrete, conjectural), a grand conjecture or major hypothesis, and theoryA (abstract, analytical), a body of abstract principles; (2) separate theories from spin; and (3) consider the relationships of persons to theories.
5.2. The Modern Synthesis of 1959. This post describes and documents a cohesive theory of evolutionary genetics that has been quietly abandoned, though it still remains cryptically influential.
5.3. The abuse of history.This post explores (1) Synthesis Historiography, a set of myths about alternative theories and historical developments fabricated mainly by Mayr, and (2) the unconscious introduction of distortions by normality drift and back-projection.
5.4. Synthesis apologetics. Synthesis apologetics is not a coherent scientific position, but set of rhetorical tactics. One tactic is to craft a broad description of the Synthesis that is historically correct in what it includes, but deceptive in what it leaves out.
Weekend reads: Researcher resigns following questions about ties to China; grad student’s suicide sparks misconduct investigation; study of chronic fatigue syndrome corrected
Posted by weekend readsin