Jul 21 2018

Book review: Sentimental Savants by Meghan K. Roberts

sentimental_savantsSentimental Savants by Meghan K. Roberts is about the role of the families of savants in the Enlightenment. In this respect it covers some of the themes of Patricia Fara’s book, Pandora’s Breeches.

The book focuses on French families, the author makes clear that this isn’t because of any special feature of French families, rather that one has to start somewhere. Each chapter covers a different facet of family life.

As usual for writing about this time I will clumsily switch between savant, natural philosopher and possibly even the anachronistic “scientist”. You know who I mean!

Prior to the Enlightenment the work of the “philosophe” was associated with monastic traditions of celibacy, or at the very least the absence of a married partner. This tradition maintained, in at least the University of Cambridge, for many years. Cambridge fellows were finally allowed to marry around 1860. During the Enlightenment in France, and elsewhere the personal life of the savant started to become important, being a good family man was seen as a benefit, it gave a savant moral authority. The evidence for these changes can be seen in the changes in style of the eulogies written for savants across the 18th century, publications including plays and learned works and through correspondence.

Women started to play a visible role in scientific and intellectual discourse during the Enlightenment. They did not have the full freedom of study that their male counterparts had. They were seen as record keepers, promoters, writers, translators, and housekeepers. It wasn’t unusual for a wife, or daughter, to be trained to the role that the man of the house had decided for her. The family was seen as part of a demonstration of savant as well-rounded person, women started to be seen as intellectuals in their own right. Émilie du Châtelet falls very clearly into this class.

In the 18th century inoculation against smallpox became a topic of public debate. Inoculation is a little different from vaccination, the patient is treated with the live virus and gets at least symptoms of the disease (a vaccine uses a dead microbe, or part of it). This makes it more risky than vaccination. In 18th century France philosophes would inoculate their children, and write publically about what they had done. This was to give moral authority to their arguments. A savant inoculating his children would be showing that he had thought about the risks and had gone ahead with the inoculation for the good of his child and of society.

The education of children is a theme of a further chapter. Rousseau makes an early entry in the book on the occasion of leaving his five children at the foundling hospital! Other savants took more care of their children. In France prior to 1762 much education was managed by the Jesuits but in 1762 they were banned from teaching. This left a gap which savants tried to fill. Although intellectual parents sought to educate both sons and daughters they approached this in a gendered fashion. Typically boys were prepared for prestigious roles in society, girls were prepared to be good wives.

The Lavoisiers make a couple of appearances in the book, first in describing the role of Marie-Anne Lavoisier in Antoine’s work and later in a chapter on Antoine’s work on his farm. Antoine was “father” to his estate, not only was he seeking to improve agricultural techniques for the benefit of France. He also wished to be seen as a paternal figure to the community, supporting it in times of need, helping to resolve disputes and tending to the sick.

The book does not present a glorious revolution whereby women came fully into the scientific community, frankly this hasn’t really happened yet – certainly in some fields – but it does show how the role of savant / academic / scientist started to move away from the lonely genius to a man embedded in family life. From the point of view of women, it represented a time where women started to become, explicitly, part of the scientific enterprise.

I notice that Roberts cites Ken Alder in her acknowledgements, his book (The Measure of All Things) on measuring the size of the Earth by triangulating a meridian through revolutionary France is well worth reading – as is Poirier’s book Lavoisier: Chemist, Biologist, Economist. As a testament to the power of blogging I see that Poirier also wrote a biography of Marie-Anne Lavoisier (La science et L’amour: Mme Lavoisier), mentioned in the comments to my post of his book on Antoine.

I found this an enjoyable read, introducing a facet to the history of science that is relatively little covered.

Jun 27 2018

Book review: The Devil’s Doctor by Philip Ball

devils_doctorThe Devil’s Doctor by Philip Ball is a biography of Paracelsus, and a view of Renaissance magic and science. Paracelsus, who lived from 1493 to 1541 was a contemporary of figures such as Copernicus (1473-1543) and Vesalius (1514-1564), who preceded closely William Harvey, William Gilbert,Galileo and Francis Bacon.

Each chapter says something about Paracelsus and something about the time in which he lived. So we start with his early life growing up in a small town in Switzerland, his father a physician. Before moving on to a town in what is now Austria, where his father worked as a teacher in a mining school. The aside here is on the early mining industry, in this case in the Austrian Tyrol, dominated by the entertainingly named Fugger family who had gained their pre-eminent position by lending money to the nobility with silver mines as collateral.

Mining, or rather the refining of ores into metal is a “scientific” process, mixtures of seemingly unpromisingly materials are heated and, as if by magic, a shiny and valuable metal appears. In the 16th century we would not recognise the theories by which the refining process was rationalised but they highlight the importance of understanding what was to become chemistry for profit.

What’s notable about these two chapters is that one can easily see how Paracelsus started learning early. Subsequent chapters cover Paracelsus’s own education as a physician, although it is not clear whether he ever “qualified”, and a tour of Europe chased by war.

Paracelsus was a contemporary of Martin Luther and on the face of it they and something in common in their desire to overthrow the status quo. However, Luther was very focussed on solving what he saw to be the shortcomings of the administrative side of the Catholic church. Fundamentally, he wasn’t interested in science and its reform away from the scholastic creed. Luther’s view was that God was Divine and had no need to construct a Creation amenable to scientific understanding.

Paracelsus was interested in overthrowing the status quo, particularly in medicine. To that point medicine was largely about following the teachings of the ancient Greeks whose surviving medical practice was based on the four humours. Medicine was about substances and processes to correct imbalances in the humours which arose in specific diseases. Paracelsus emphasised practical study and observation rather than cleaving to these old theoretical models for medicine. He advised seeking out folk remedies specific to places, believing that diseases and their cures were localised. So his world view was something akin to our own but not exactly so, he represents a transitional state. He also counselled only undertaking minimal surgery – exceedingly wise advice at the time.

Paracelsus’s study of “chemistry” went hand in hand with his medical work. He experimented to make medicines extracting chemicals from natural sources, combining them with other materials. At the time this type of work was usually tied up with alchemy, which was interested in turning base materials into gold and elixirs for eternal life. Alchemy was part chemistry and part mysticism and secrets. Paracelsus practicised each of these elements although it seems he was less enamoured of the secretive side. His alchemical writings were more straightforward than some of his contemporaries and predecessors. The theoretical underpinnings he provided for his work seem bizarre to us but chemistry is a difficult thing. It took hundreds of years to reach the point at which we are now and Paracelsus came early in that time.

Local man (to me) Robert of Chester gets a mention as a translator of alchemical texts. Authorship seemed a fluid concept in the early period with some people attributing their work to more renowned writers – difficult to disentangle after so many years.

For the chemists amongst us, it seems that Paracelsus discovered diethyl ether whilst trying to make sulphuric acid palatable by mixing it with wine. It is mindboggling that sulphuric acid was used as a medicine!

Many of the chapters end with Paracelsus having fled a city, typically because he had offended one authority or another. He was certainly cantankerous, it is difficult to say whether he was personally unpleasant. The way he practiced medicine was, to a degree, out of tune with the times and seemed more caring than his contemporaries. I was struck by how well documented Paracelsus’s life seemed to be, perhaps this is partly a result of his peripatetic nature.

The book finishes with a couple of chapters set after Paracelsus’s death, the first of these looks at how his writings were collected and published after his death by a band of supporters, and how a weaker band of opponents also wrote about him. This conflict probably also contributed to the relatively high level of documentation for his life. The final chapter looks at his longer term impact, this is a mixed picture. Unlike Copernicus, Vesalius, Harvey and the like he did not leave discoveries which are valid today. Rather his methodology informed the changes that were to take place in the 17th century.

May 19 2018

Book review: The Anatomy of Colour by Patrick Baty

anatomy_of_colourThe Anatomy of Colour by Patrick Baty is a history of painting as decoration for houses and buildings stretching back 350 years or so. The Painter’s Company dates all the way back to 1283, and the practice of decorative house painting back into antiquity. There is direct evidence for this preserved in Pompeii.

As I recall I’ve been following Patrick on social media for quite some time, in part because what he does relates to my former professional interests. Anatomy picks up on work I used to do in colour measurement. I did it in a different context – for a “fast moving consumer goods” company making washing liquid, shampoo and the like. It also has some relevance to work I do now on dating buildings.

At the beginning of the book the materials used to generate colours and paints are discussed. Typically these are minerals or plant materials, synthetic colourants only started to become available in the 19th century. Lead carbonate (white lead) was long used as the basis for many oil paints (of all colours), despite it’s known toxicity. It was replaced by zinc oxide in the later 19th century and now titanium dioxide is used. Confusingly from the early period “pink” used to mean a yellowish colour derived from plant material. Also mentioned are the linseed oil that formed the carrier for most paints well into the 20th century.

Following this introduction a number of broad themes are discussed in successive periods so for example in each period we learn about the favoured colours and colour schemes, colour theories and systems, and the key books relating to decorating and colour. Technology is a slowly changing background to this. In the beginning painters bought the pigments and oils and made their own paint, possible making pigments from raw materials. Pigment pastes in metal tubes came into use in the 19th century with tinned paints a 20th century innovation, as far as I can tell.

I’ve always had a problem with distemper, a suspension of chalk in a glue base which can be washed off and reapplied, mainly because I learnt about distemper, the disease of dogs first! Distemper was used as a disposable wall covering until the late 19th century when it started to be replaced by other commercial formulations until they were in turn replaced with emulsion paints sometime after the Second World War.

Something I hadn’t considered before was the importance of colour in horticulture and zoology. Darwin took a book on colour on his trip around the world. He needed it to describe the colours of animals and plants as they were collected since there was no photography and specimen preservation techniques would not necessarily preserve colour. More recently colour systems have been developed around the needs of horticultural. The red of a robins breast is probably as good a reference to a colour as could be obtained artificially until into the 20th century. Similar the yellow of a daffodil.

The colour systems discussed are a little different from those I used in scientific colour measurement, the closest approach is the Munsell “hue, saturation, value” system. In general the systems presented here are focused around defining words to describe colours to aid communication and specification, and establishing harmonious colour combinations. My work was more involved in measuring colour in a machine to see the effect of different washing liquid formulations, or similar.

In the 20th century Britain saw standardisation of colours with recommendations for the painting of commercial and public buildings. The post-war rise in DIY is well-known but housepainting was seen as something a gentleman might undertake even in the 18th and 19th century.

Early on in the book Baty shows some paint cross-sections which are discussed very briefly. I must admit I find this forensic side very interesting and I was a bit disappointed that there was not more of this – I can see how it doesn’t fit with the main audience for the book. I found the chapter / section numbering a bit confusing, there are two levels in the hierarchy and both use Roman numerals!

The book is beautifully illustrated with colour reproductions of many of the different colour systems used over the years, as well as photographs of interiors following the styles discussed. For practioners in the investigation (and recreation) of decorative schemes I can see this book being absolutely essentially, and as a more casual reader I enjoyed it too.

May 13 2018

Book review: Other minds by Peter Godfrey-Smith

other_mindsOther Minds by Peter Godfrey-Smith is about consciousness through the lens of cephalopods, a group comprising octopuses, cuttlefish, squid and the nautiluses.

Godfrey-Smith is a philosopher, rather than a scientist. This reflects the theme of the book, Godfrey-Smith’s idea is to understand our consciousness by looking at a creature with as radically different a consciousness as he can find. So his book is more a philosophical rather than a scientific view on consciousness which to my mind is no bad thing.

I was particularly impressed by Charles the octopus, one of three which experimenters used to try to measure octopus intelligence. The other two octopuses in their study put in some effort to carry out the tests presented whilst Charles insisted on squirting water at the experimenters and being otherwise uncooperative. It does make you wonder whether measures of animal intelligence are more a combination of willingness and intelligence. Perhaps the smarter animals just can’t be doing with intelligence tests. Squirting water at lights in aquariums to put them out seems, apocryphally, to be quite common behaviour amongst octopuses.

Mrs H is undertaking a doctorate in education and it strikes me that her mode of doing research, qualitative with relatively few subjects has more in common with this type of study than the experimenters care to admit. In this type of experiment the narrative rather than the result may be more important. So measuring the intelligence of the octopuses, or the time it took them to complete a learning task, is less important than the narrative of how they performed in the experiment.

The book weaves together cephalopod biology with Godfrey-Smiths own observations of cephalopods in the wild. I was intrigued to learn that the octopus brain is wrapped around the digestive tract which has been observed to lead to problems when attempting to eat particularly spikey foods. More generally, the neural material of an octopus is not all to be found in the “brain” it is distributed around the body. Octopus legs appear to have a degree of autonomy in at least deciding how to achieve a goal, even if the “brain” decides what the end goal might be.

Cuttlefish and octopuses have amazing abilities to change their appearance both in terms of colour, but also physical texture – they can make themselves bumpy. Interestingly, as far as we can tell they are unable to see the colours they produce – the photoreceptors in their eyes are of a single type. However, there is clearly something else going on – photoreceptors are found throughout the body of the cephalopod, and they are able to match background colours exquisitely. Furthermore, with the chromatophore and other colour producing structures also present a different mechanism for producing colour sensitivity is quite feasible – for example using those chromatophores as a filter “wheel” that sits in front of the light sensitive cells.

Godrey-Smith highlights that, although the cephalopods have a huge ability to make signals, because they are not a social species their ability to use those signals is limited. This is contrast to species such as baboons who have a more limited ability to generate vocal signals but, as a social species, have a much greater ability to interpret those signals in terms of establishing their position in a hierarchy and understanding how the hierarchy has changed.

Close to the end of the book, there is a terrible denouement: the typical cephalopod lifespan is only a couple of years. These creatures, so full of potential, are but brief inhabitants of the earth. Godfrey-Smith draws parallels with the Replicants in Bladerunner here. Their brief lives are understood in terms of their dangerous environment which has led to an evolutionary strategy of large broods of young, easily lost.

The book finishes with a discussion of “Octopolis” a location in Australia where octopuses, unusually, congregate and where Godfrey-Smith did a large part of his observations discussed throughout the book. Also we find here that he is involved in scientific publication.

Overall, one is left with the feeling of cephalopods having been an opportunity missed in the consciousness stakes. They have all the mental machinery but their truncated lives and limited social behaviour means that in all likelihood the opportunity is unfulfilled. This is a consciousness-centric human view, no species is striving for consciousness or intelligence they are doing what is needed for there species to continue in the niches they find themselves.

Apr 26 2018

Book review: Hands-on machine learning with scikit-learn & tensorflow by Aurélien Géron

machine-learningI’ve recently started playing around with recurrent neural networks and tensorflow which brought me to Hands-on machine learning with scikit-learn & tensorflow by Aurélien Géron, as a bonus it also includes material on scikit-learn which I’ve been using for a while.

The book is divided into two parts, the first, “Fundamentals of Machine Learning” focuses on the functionality which is found in the scikit-learn library. It starts with a big picture, running through the types of machine learning which exist (supervised / unsupervised, batched / online and instance / model) and then some of the pitfalls and problems with machine learning before a section on testing and validation. The next part is a medium sized example of machine learning in action which demonstrates how the functionality of scikit-learn can be quickly used to develop predictions of house prices in California based on census data. This is a subject after my own heart, I’ve been working property data in the UK for the past couple of years.

This example serves two purposes, firstly it demonstrates the practical steps you need to take when undertaking a machine learning exercise and secondly it highlights just how concisely much of it can be executed in scikit-learn. The following chapters then go into more depth first about how models are trained and scored and then going into the details of different algorithms such as Support Vector Machines and Decision Trees. This part finishes with a chapter on ensemble methods.

Although the chapters contain some maths, their strength is in the clear explanations of the methods described. I particularly liked the chapter on ensemble methods. They also demonstrate how consistent the scikit-learn library is in its interfaces. I knew that I could switch algorithms very easily with scikit-learn but I hadn’t fully appreciated how the libraries generally handled regression and multi-class classification so seamlessly.

I wonder whether outside data science it is perceived that data scientists write their own algorithms from scratch. In practical terms it is not the case, and hasn’t been the case since at least the early nineties when I started data analysis which looks very similar to the machine learning based analysis I do today. In those days we used the NAG numerical library, Numerical Recipes in FORTRAN and libraries developed by a very limited number of colleagues in the wider academic community (probably shared by email attachment).

The second part of the book, “Neural networks and Deep Learning”, looks at the tensorflow library. Tensorflow has general applications for processing multi-dimensional arrays but it has been designed with deep learning and neural networks in mind. This means there are a whole bunch of functions to generate and train neural networks of different types and different architectures.

The section starts with an overview of tensorflow with some references to other deep learning libraries, before providing an introduction to neural networks in general, which have been around quite a while now. Following this there is a section on training deep learning networks, and the importance of the form of activation functions.

Tensorflow will run across multiple processors, GPUs and/or servers although configuring this looks a bit complicated. Typically a neural network layer can’t be distributed over multiple processing units.

There then follow chapters on convolution neural networks (good for image applications), recurrent neural networks (good for sequence data), autoencoders (finding compact representations) and finally reinforcement learning (good for playing pac-man). My current interest is in recurrent neural networks, it was nice to see a brief description of all of the potential input/output scenarios for recurrent neural networks and how to build them.

I spent a few years doing conventional image analysis, convolution neural networks feel quite similar to the convolution filters I used then although they stack more layers (or filters) than are normally used in conventional image analysis. Furthermore, in conventional image analysis the kernels are typically handcrafted to perform certain tasks (such as detect horizontal or vertical edges), whilst neural networks learn their kernels in training. In conventional image analysis convolution is often done in Fourier space since it is more efficient and I see there experiments along these lines with convolution neural networks.

Developing and training neural networks has the air of an experimental science rather than a theoretical science. That’s to say that rather than thinking hard and coming up with an effective neural network and training scheme one needs to tinker with different designs and training methods and see how well they work. It has more the air of training an animal the programming a computer. There are a number of standard training / test sets of images and successful models trained against these by other people can be downloaded. Such models can be used as-is but alternatively just parts can be used.

This section has many references to the original literature for the development of deep learning, highlighting how recent this new iteration of neural networks is.

Overall an excellent book, scikit-learn and tensorflow are the go-to libraries for Python programmers wanting to do machine learning and deep learning respectively. This book describes their use eloquently, with references to original literature where appropriate whilst providing a good overview of both topics. The code used in the book can be found on github, as a set of Jupyter Notebooks.