Language evolution in action

Here’s an example of how a word can completely reverse its meaning, from a BBC article about the Mid-Levels escalator in Hong Kong (also a cool subject in itself):

“We did the planning in 1984,” says architect Remo Riva, director of P&T Group, which consulted the government.

The verb “consult”, when used in a sentence like “A consulted B”, usually means that A asked B for their opinion. However in this case, it’s the P&T Group which were offering advice to the Hong Kong government, and not the other way around. So the sense of the word “consult” appears to have evolved like this:

“The government consulted the P&T Group” → “The P&T Group were consultants for the government” → “The P&T Group consulted for the government” → “The P&T Group consulted the government”

Funny to see this in action. And yet it’s still very strange to imagine that the word “nice”, for example, used to be pejorative, meaning a foolish or silly person.

Or in the case of the word “sanction”, made me completely misinterpret this headline: “Thai people urged to ‘socially sanction’ critics of monarchy”.

Binning trees by topology

Recently stumbled across a 2013 paper from Ryan and Irene Newton describing a tool, called PhyBin, for binning phylogenetic trees, i.e. clustering them by similarity into groups (“bins”). They use the Robinson Foulds metric to represent the distance between trees.

The reason for doing this is to look at the phylogenies of individual gene ortholog clusters in a set of genomes, to find those genes that have a phylogeny different from the others. This might be useful e.g. to detect genes that have undergone horizontal gene transfer. The example they used for their paper was the insect symbiont Wolbachia.

It seems like a nice way to screen a set of genomes for genes that might be interesting. I had wanted to try to do something like this, but with a concordance-factor approach instead. Some other thoughts:

  • Each gene is represented by one tree – uncertainty is not taken into account, unlike with concordance factors, as implemented in BUCKy for example
  • If there are horizontally-transferred genes, they would probably have patchy distribution and not be in every species. But such genes that are present in only some genomes would be pre-excluded from the analysis, also in concordance analysis. In PhyBin paper the authors mention the case of Wolbachia prophage which has precisely this limitation.
  • Collapsing short branches is a good idea

Lonely negatives

My sister and I are great fans of the British political comedy series Yes, Minister and its sequel Yes, Prime Minister. A joke from the episode “The Greasy Pole” that she recently reminded me about:

Joan Littler: What does “inert” mean?

Sir Humphrey: Well it means it’s not… ert.

Bernard: [to himself] Wouldn’t ert a fly.

(via Wikiquote)

I’ve been keeping a little list of words like “inert”, that appear to be a negative form, but whose positive partners aren’t in common use. I’m not sure if there’s already a term for them, but I’d like to call them “lonely negatives”. They’re a curious crowd; here I dip into an etymological dictionary (Concise Oxford Dictionary of English Etymology, ed. T. F. Hoad) and attempt to make a taxonomy of these rogues….

Continue reading

Biography of Seah Eu Chin now online

Many moons ago, I researched and wrote up a biography of my ancestor Seah Eu Chin (1805-1883). He was a pioneer in the agricultural development of Singapore and Johor, and also one of the founders of the Teochew organisation the Ngee Ann Kongsi. The draft has been sitting around for a long time, and I have now finally converted it to a website,

Working on his biography has stimulated my interest not only in Singapore history, but also in the broader fields of economic and social history. I hope that the way I present his story does not seem like hagiography, but instead puts his life into the context of the times that he lived in. Like a fish oblivious to the water it swims in, we tend not to think too much about the origins of the social order we live in and the institutions embedded in it. The 19th century seems irrelevant to most Singaporeans, and we only learn the barest outlines of the political history of this period in school. The history of immigration, for example, is often depicted as a dichotomy between faceless coolies on one hand, and the rags-to-riches stories of rich merchants like Seah Eu Chin on the other.

In fact, echoes of the 19th century are still with us. Some of these are indeed forgotten corners waiting to be rediscovered – the tomb of Seah Eu Chin was long thought to be lost and was only found again in a wooded area of Toa Payoh in 2012. But other traces are right in front of us just waiting to be seen:

Why do we have several place names in the north of the island that end in “Chu Kang”? This is a relict of gambier agriculture that used to be organized around river or inlet “harbours”, from which the goods and provisions could be loaded. These plantations were not just on Singapore island but also along the Johor coast, and boat traffic connected the Johor plantations with the merchants, like Seah Eu Chin, in the port of Singapore.

Why is one of the biggest malls in Orchard Road called Ngee Ann City? What business does a charity have in owning a shopping centre? It turns out that the site used to be a Teochew cemetery, run by the Ngee Ann Kongsi on land that it bought in 1845, and Seah Eu Chin was one of the original trustees.

Many issues that were once argued over in private homes, at clan organizations, and the legislative chamber are still with us today. I found it remarkable to read in the archives about disputes over the regulation of gambling, anxiety about immigration, and potential tax evasion. Of course the details may differ – people now bet illegally online and not in shady gambling dens, convict labourers are no longer a key source of manpower for public works, and opium is not subject to tax – but it is still surprising how many parallels remain.

In retrospect I should have done this sooner. With content on the web, I can easily update the pages with new information, and embed media including maps and videos. I hope that this work will be interesting to more than just myself and other Seah descendants, and that it will get at least some readers interested in connecting our present to the past.

What your favorite Bacon says about you

Classic pan-fried


You appreciate the good things in life, and you know that they don’t have to be particularly expensive or good-looking to be tasty. You have a hedonistic side; while you may know all the reasons that fried bacon is bad for you, it still doesn’t stop you from indulging in it once in a while.


You enjoy movies, and also enjoy thinking about them. You could carry your own in a conversation about your favorite actor, who is probably not an A-list superstar but definitely has acting chops. For you, an actor’s career arc can be as interesting as the actual movies they act in. Your favorite math class was graph theory.



Statue of Roger Bacon in the Oxford University Museum of Natural History. Photograph taken by Michael Reeve (en:User:MykReeve), 30 May 2004.

People tend not to notice you the first time around, and if they do it’s usually because of some folly you made that you don’t take too seriously yourself, and this can be frustrating for you. You are a rational person, but faith is also a very important part of your life.


388px-Francis_BaconThe life of the mind appeals to you, but you are too much of a realist to ignore the real world. You are not afraid to get your hands dirty, figuratively or literally speaking. You need a better winter jacket.


You don’t actually like bacon, do you?

Plotting data as ratios? Think again

We are often interested in ratios between two quantities. As an example, let’s use data from a study on the sugar content of soft drinks, where the the sugar content declared on the drink label was compared to the actual sugar content measured in the laboratory (Ventura et al. 2010, Obesitypdf). The paper includes a nice table summarizing their measurements, which I have adapted to produce the plots shown here.

How can we present this data to get the most insight? In my opinion, presenting such data as ratios can obscure useful information; showing scatterplots of the two quantites can make it easier to spot patterns.
Continue reading

New gbtools release (v2.5.2)

Instead of working I’m procrastinating by fixing bugs in my software. The new version of gbtools includes two new features that improve the plotting of taxonomic markers, and some fixes to long-standing bugs. You can now adjust how many taxa are colored and included in the plot legend (thereby avoiding cluttered plots with too many colors to interpret), and also highlight individual taxa.

Wondering what gbtools is? Read my previous blog post, or the paper published last December.