Opinion

13th Amendment, fair political representation, and social-choice theory

Published

2 years ago

2024/06/25

Sahan Weerasooriya

The signing of the Indo-Lanka Accord, which led to the establishment of Provincial Councils.

by Chandre Dharmawardana,
chandre.dharma@yahoo.ca

The TNA is the main political party of the North. S. Shritharan was recently elected its leader and M. A. Sumanthiran, who is regarded by some as being “barely Tamil”, as one Eelamist resident in Canada put it, was sidelined. Sritharan’s vision, expressed in post-election speeches, demands the merger of the Northern and Eastern provinces; he rejects the 13A as being grossly inadequate to meet the aspirations of the Tamils. The political parties of Gajendra Ponnamblam, and of C. V. Wigneswaran takes an even harder public stand. All tactically reject 13A, even though they rush to India to support 13A when support for 13A weakens in the South. The positions taken by southern politicians regarding 13A are also merely tactical and opportunistic.

Ironically, 13A is already a part of Sri Lanka’s Constitution, with some parts of it implemented, and others in suspense, mainly due to a huge lack of trust across the Northern and Southern political formations. Even the Eastern Tamil leaders do not trust the Northern leaders.

While the minority leaders still seek the chimera of an Indian supervisory role, the majority-community politicians know that strong Indian interventions, even “parippu dropped from air” are no longer a part of the show. President Ranil Wickremasinghe was seated next to Prime Minister Narendra Modi at the latter’s inauguration, while no TNA leader was visible. Meanwhile, the provincial councils themselves have atrophied, with provincial elections not even considered worth the cost, under the current circumstances.

The Northern political leaders rightly believe that any government in Colombo will be a government of the Majority Community and that minority rights will NOT be protected under such a set-up, judging by past history. So, they aspire to have a separate government of their own as the “only effective approach”. However, this approach triggered the past history of communal politics and violence that led to terror and counter-terror. Finally, the TULF leaders, Sinhalese politicians, even the Indian Leader who fathered the 13A, and thousands of innocent civilians got wiped out.

If there is no trust, there can be NO federalism, nor an effective 13A. Even an independent Eelam, separate from a Sinhalé by a physical border is not viable, as the two neighbours will be continually at war, as is the case between India and Pakistan, or across and even within Indian states (e. g. Manipur), even though the “Indian Model”, like 13A, is claimed to resolve these conflicts. Furthermore, such “independent” states will be forced to join up with big powers and become mere pawns of global proxy wars. That is the end of their “self-determination”.

The TNA says, “We don’t trust the majority, so we want our own government; but the minorities who will be under us, i.e., Muslims of the East or any Sinhalese who live in our “exclusive homeland” must trust us. Just forget attacks on Muslims or Sinhalese minorities when the TNA was an LTTE proxy”! This “aspiration” for hegemony by Tamil leaders over other minorities will be rejected by the respective minorities, just as the Tamil leaders reject being ruled by the Majority that they do not trust.

Social-choice theory

How can we equitably allocate agents (or electoral seats) to represent a group of people within a unitary setup (with a total quota of 225 seats), or with subdivided setups (e. g., with provincial councils or federal states with quotas of seats reflecting minority groups)?

This question falls within a class of much studied mathematical problems in game theory, mathematical economics as well as in the theory of social choice. Intellectual giants like John von Neumann and other mathematicians pioneered these studies. However, the most important results relevant to our discussion here came from Blinski and Young as well as from Kenneth Arrow. The latter won the Nobel Prize for economics in 1972 for his theorems on “social-choice theory”.

Blinski and Young proved a theorem showing that any apportionment rule (or representation and devolution rule) that stays within an assigned quota (say, of seats) suffers from what is known as the population apportionment paradox. This states that unless the populations remain absolutely static, even if the minority has a decisively large rate of population growth, the majority still gains more representation (or more power) inexorably! There is NO fair apportionment scheme!

Blinsky and Young’s result was a surprising “no-go” theorem. However, Arrow’s theorem, formulated in 1951 was even more surprising and counter-intuitive. Arrow laid down five “self-evident” axioms (or rules) about what may be called the “Will of the People” to be represented. For instance, a key rule is that the preferences and aspirations of a group should be chosen only from the group members (and not from outsiders). Another axiom is that the “will of the group” must not be that of one particular person; this is known as the no-dictator rule. The other axioms are similar harmless-looking rules about the group having specific preferences (e.g., favouring a set of religious or cultural traits against another set), or having maverick members who have changed policies in the past on a specific preference, although now in accordance with the “will of the group”.

Arrow’s impossibility theorem

Kenneth Arrow proved that, in spite of the highly democratic and seemingly “fair” formulation of these axioms, no such fair representation is possible. This is known as Arrow’s Impossibility Theorem. This theorem states that mandating the preferences and aspirations of the group cannot be ensured while adhering to usual “democratic” principles of fair voting procedures!

The mathematical conclusion is that a selection of people making decisions for those who elected them can never be a rational or fair process, however wise or benevolent they are! Their decisions will be necessarily autocratic! Naturally, the minorities within any group, be it under the Sinhalese majority in the main government, or under the Tamil majority in the TNA government reigning over the North and the East, will discriminate against the minority in each case.

Every available constitutional representation that satisfies Arrow’s axioms (i.e., common-sense ideas of fairness) is a perverse one. There is no “will of the people” or a democratic way of representing it. This very painful conclusion, reached by mathematicians in the 1950s, has stood all critical attacks on it. For over twenty-two decades, political scientists for whom the concept of the “will of the people” is as sacrosanct as the geocentric universe was to the medieval church attacked it! Instead of disproving Arrow, similar impossibility theorems, no-cloning theorems, etc., have been established in quantum information theory and quantum mechanics.

Devising an electoral scheme is mathematically equivalent to an apportionment scheme. Instead of allocating seats on the basis of population (i.e., “The People”), one may consider allocating “seats” on the basis of votes. This leads to models based on proportional representation (PR) instead of apportionment.

Mathematicians have shown that PR leads to even more serious negative consequences than apportionment. A variety of paradoxes of the Blinsky and Young type have been established. A very serious conclusion is that even the mildest PR system will confer a disproportionate amount of power to the third largest party in parliament! The third largest party becomes the king maker and often comes into a coalition with the second-ranking party to become the government! The validity of these results from game theory in practical politics has been established by studies of the history of governments in Germany, Israel and Denmark where high levels of proportional government have been legislated.

In my opinion, a way around these problems is to abandon electoral methods and return to the method of SORTITION advocated by Aristotle and used in several Hellenic cities during the time of Pericles.

Sortition has been adopted today in various limited ways, especially for local or provincial governments, in Ireland, France, Belgium, Canada and even Mongolia. In the simplest sortition model one arbitrarily selects by lottery a group of people who constitute the parliament. While these legislators last only five or six years, it is the administrative service that persists. The sortition parliament is not claimed to represent the “will of the people”. The lottery may be open to all the people, or only to a selection defined by their public service, education etc., as specified by a parliament chosen initially by simple sortition. That is, the first sortition parliament may enact more elaborate sortition models, but ensuring that the random element implied by sortition is never negated.

The sortition model ensures that the same set of corrupt politicians do not continue to get elected every time by controlling the list of candidates as well as the vote-gathering infrastructure which favours existing parties that have accumulated much wealth, by hook or crook. It also eliminates demagogues as the election is by lottery.

In other words, SORTITION ensures that a “system change” occurs every time. It ensures that political crooks, their henchmen and progeny do not entrench themselves and hold onto power over decades and decades, be it in the North or the South. I had given a discussion of the sortition model in a previous article in the Island (02-01-2023). It may also be accessed via the web (https://thuppahis.com/2023/01/02/crunchtime-resolving-sri-lankas-political-dilemma/ The applicability of the sortition model to the political problems in the USA has been discussed in the Harvard Review of politics (https://harvardpolitics.com/sortition-in-america/).

All News Advertisement

Opinion

Capt. Dinham Suhood flies West

Published

18 hours ago

2026/02/13

Thilina Walpola

A few days ago, we heard the sad news of the passing on of Capt. Dinham Suhood. Born in 1929, he was the last surviving Air Ceylon Captain from the ‘old guard’.

He studied at St Joseph’s College, Colombo 10. He had his flying training in 1949 in Sydney, Australia and then joined Air Ceylon in late 1957. There he flew the DC3 (Dakota), HS748 (Avro), Nord 262 and the HS 121 (Trident).

I remember how he lent his large collection of ‘Airfix’ plastic aircraft models built to scale at S. Thomas’ College, exhibitions. That really inspired us schoolboys.

In 1971 he flew for a Singaporean Millionaire, a BAC One-Eleven and then later joined Air Siam where he flew Boeing B707 and the B747 before retiring and migrating to Australia in 1975.

Some of my captains had flown with him as First Officers. He was reputed to have been a true professional and always helpful to his colleagues.

He was an accomplished pianist and good dancer.

He passed on a few days short of his 97th birthday, after a brief illness.

May his soul rest in peace!

To fly west my friend is a test we must all take for a final check

Capt. Gihan A Fernando

RCyAF/ SLAF, Air Ceylon, Air Lanka, Singapore Airlines, SriLankan Airlines

Opinion

Global warming here to stay

Published

18 hours ago

2026/02/13

Thilina Walpola

The cause of global warming, they claim, is due to ever increasing levels of CO2. This is a by-product of burning fossil fuels like oil and gas, and of course coal. Environmentalists and other ‘green’ activists are worried about rising world atmospheric levels of CO2. Now they want to stop the whole world from burning fossil fuels, especially people who use cars powered by petrol and diesel oil, because burning petrol and oil are a major source of CO2 pollution. They are bringing forward the fateful day when oil and gas are scarce and can no longer be found and we have no choice but to travel by electricity-driven cars – or go by foot. They say we must save energy now, by walking and save the planet’s atmosphere.

THE DEMON COAL

But it is coal, above all, that is hated most by the ‘green’ lobby. It is coal that is first on their list for targeting above all the other fossil fuels. The eminently logical reason is that coal is the dirtiest polluter of all. In addition to adding CO2 to the atmosphere, it pollutes the air we breathe with fine particles of ash and poisonous chemicals which also make us ill. And some claim that coal-fired power stations produce more harmful radiation than an atomic reactor.

STOP THE COAL!

Halting the use of coal for generating electricity is a priority for them. It is an action high on the Green party list.

However, no-one talks of what we can use to fill the energy gap left by coal. Some experts publicly claim that unfortunately, energy from wind or solar panels, will not be enough and cannot satisfy our demand for instant power at all times of the day or night at a reasonable price.

THE ALTERNATIVES

It seems to be a taboo to talk about energy from nuclear power, but this is misguided. Going nuclear offers tried and tested alternatives to coal. The West has got generating energy from uranium down to a fine art, but it does involve some potentially dangerous problems, which are overcome by powerful engineering designs which then must be operated safely. But an additional factor when using URANIUM is that it produces long term radioactive waste. Relocating and storage of this waste is expensive and is a big problem.

Russia in November 2020, very kindly offered to help us with this continuous generating problem by offering standard Uranium modules for generating power. They offered to handle all aspects of the fuel cycle and its disposal. In hindsight this would have been an unbelievable bargain. It can be assumed that we could have also used Russian expertise in solving the power distribution flows throughout the grid.

THORIUM

But thankfully we are blessed with a second nuclear choice – that of the mildly radioactive THORIUM, a much cheaper and safer solution to our energy needs.

News last month (January 2026) told us of how China has built a container ship that can run on Thorium for ten years without refuelling. They must have solved the corrosion problem of the main fluoride mixing container walls. China has rare earths and can use AI computers to solve their metallurgical problems – fast!

Nevertheless, Russia can equally offer Sri Lanka Thorium- powered generating stations. Here the benefits are even more obviously evident. Thorium can be a quite cheap source of energy using locally mined material plus, so importantly, the radioactive waste remains dangerous for only a few hundred years, unlike uranium waste.

Because they are relatively small, only the size of a semi-detached house, such thorium generating stations can be located near the point of use, reducing the need for UNSIGHTLY towers and power grid distribution lines.

The design and supply of standard Thorium reactor machines may be more expensive but can be obtained from Russia itself, or China – our friends in our time of need.

Priyantha Hettige

Opinion

Will computers ever be intelligent?

Published

2 days ago

2026/02/12

Sahan Weerasooriya

Alan Turin and the Turin machine

The Island has recently published various articles on AI, and they are thought-provoking. This article is based on a paper I presented at a London University seminar, 22 years ago.

Will computers ever be intelligent? This question is controversial and crucial and, above all, difficult to answer. As a scientist and student of philosophy, how am I going to answer this question is a problem. In my opinion this cannot be purely a philosophical question. It involves science, especially the new branch of science called “The Artificial Intelligence”. I shall endeavour to answer this question cautiously.

Philosophers do not collect empirical evidence unlike scientists. They only use their own minds and try to figure out the way the world is. Empirical scientists collect data, repeat and predict the behaviour of matter and analyse them.

We can see that the question—”Will computers ever be intelligent?”—comes under the branch of philosophy known as Philosophy of Mind. Although philosophy of mind is a broad area, I am concentrating here mainly on the question of consciousness. Without consciousness there is no intelligence. While they often coincide in humans and animals, they can exist independently, especially in AI, which can be highly intelligent without being conscious.

AI and philosophers

It appears that Artificial Intelligence holds a special attraction for philosophers. I am not surprised about this as Al involves using computers to solve problems that seem to require human reasoning. Apart from solving complicated mathematical problems it can understand natural language. Computers do not “understand” human language in the human sense of comprehension; rather, they use Natural Language Processing (NLP) and machine learning to analyse patterns in data. Artificial Intelligence experts claim certain programmes can have the possibility of not only thinking like humans but also understanding concepts and becoming conscious.

The study of the possible intelligence of logical machines makes a wonderful test case for the debate between mind and brain. This debate has been going on for the last two and a half centuries. If material things, made up entirely of logical processes, can do exactly what the brain can, the question is whether the mind is material or immaterial.

Although the common belief is that philosophers think for the sake of thinking, it is not necessarily so. Early part of the 20th century brought about advances in logic and analytical philosophy in Britain. It was a philosopher (Ludwig Wittgenstein) who invented the truth table. This was a simple analytic tool useful in his early work. But this was absolutely essential to the conceptual basis of early computer science. Computer science and brain science have developed together and that is why the challenge of the thinking machine is so important for the philosophy of mind. My argument so far has been to justify how and why AI is important to philosophers and vice versa.

Looking at computers now, we can see that the more sophisticated the computer, the more it is able to emulate rather than stimulate our thought processes. Every time the neuroscientists discover the workings of the brain, they try to mimic brain activity with machines.

How can one tell if a computer is intelligent? We can ask it some questions or set a test and study its response and satisfy ourselves that there is some form of intelligence inside this box. Let us look at the famous Alan Turing Test. Imagine a person sitting at a terminal (A) typing questions. This terminal is connected to two other machines, (B) and (C). At terminal (B) sits another person (B) typing responses to the questions from person (A). (C) is not a human being, but a computer programmed to respond to the questions. If person (A) cannot tell the difference between person (B) and computer(C), then we can deduce that computer is as intelligent as person (B). Critics of this test think that there is nothing brilliant about it. As this is a pragmatic exercise and one need not have to define intelligence here. This must have amused the scientists and the philosophers in the early days of the computers. Nowadays, computers can do much more sophisticated work.

Chinese Room experiment

The other famous experiment is John Sealer’s Chinese room experiment. *He uses this experiment to debunk the idea that computers could be intelligent. For Searle, the mind and the brain are the same. But he warns us that we should not get carried away with the emulative success of the machines as mind contains an irreducible subjective quality. He claims that consciousness is a biological process. It is found in humans as well as in certain animals. It is interesting to note that he believes that the mind is entirely contained in the brain. And the empirical discovery of neural processes cannot be applied to outside the brain. He discards mind-body dualism and thinks that we cannot build a brain outside the body. More commonly, we believe the mind is totally in the brain, and all firing together and between, and what we call ‘thought’ comes from their multifarious collaboration.

Patricia and Paul Churchland are keen on neuroscientific methods rather than conventional psychology. They argue that the brain is really a processing machine in action. It is an amazing organ with a delicately organic structure. It is an example of a computer from the future and that at present we can only dream of approaching its processing speed. I think this is not something to be surprised about. The speed of the computer doubles every year and a half and in the distant future there will be machines computing faster than human beings. Further, the Churchlands’, strongly believe that through science one day we will replicate the human brain. To argue against this, I am putting forward the following true story.

I remember watching an Open University (London) education programme some years ago. A team of professors did an experiment on pavement hawkers in Bogota, Colombia. They were fruit sellers. The team bought a large number of miscellaneous items from these street vendors. This was repeated on a number of occasions. Within a few seconds, these vendors did mental calculations and came out with the amounts to be paid and the change was handed over equally fast. It was a success and repeatable and predictable. The team then took the sample population into a classroom situation and taught them basic arithmetic skills. After a few months of training they were given simple sums to do on selling fruit. Every one of them failed. These people had the brain structure that of ordinary human beings. They were skilled at their own jobs. But they could not be programmed to learn a set of rules. This poses the question whether we can create a perfect machine that will learn all the human transferable skills.

Computers and human brains excel at different tasks. For instance, a computer can remember things for an infinite amount of time. This is true as long as we don’t delete the computer files. Also, solving equations can be done in milliseconds. In my own experience when I was an undergraduate, I solved partial differential equations and it took me hours and a lot of paper. The present-day students have marvellous computer programmes for this. Let alone a mere student of mathematics, even a mathematical genius couldn’t rival computers in the above tasks. When it comes to languages, we can utter sentences of a completely foreign language after hearing it for the first time. Accents and slang can be decoded in our minds. Such algorithms, which we take for granted, will be very difficult for a computer.

I always maintain that there is more to intelligence than just being brilliant at quick thinking. A balanced human being to my mind is an intelligent person. An eccentric professor of Quantum Mechanics without feelings for life or people, cannot be considered an intelligent person. To people who may disagree with me, I shall give the benefit of the doubt and say most of the peoples’ intelligence is departmentalised. Intelligence is a total process.

Other limitations to AI

There are other limitations to artificial intelligence. The problems that existing computer programmes can handle are well-defined. There is a clear-cut way to decide whether a proposed solution is indeed the right one. In an algebraic equation, for example, the computer can check whether the variables and constants balance on both sides. But in contrast, many of the problems people face are ill-defined. As of yet, computer programmes do not define their own problems. It is not clear that computers will ever be able to do so in the way people do. Another crucial difference between humans and computers concerns “common sense”. An understanding of what is relevant and what is not. We possess it and computers don’t. The enormous amount of knowledge and experience about the world and its relevance to various problems computers are unlikely to have.

In this essay, I have attempted to discuss the merits and limitations of artificial intelligence, and by extension, computers. The evolution of the human brain has occurred over millennia, and creating a machine that truly matches human intelligence and is balanced in terms of emotions may be impossible or could take centuries

*The Chinese Room experiment, proposed by philosopher John Searle, challenges the idea that computers can truly “understand” language. Imagine a person locked in a room who does not know Chinese. They receive Chinese symbols through a slot and use an instruction manual to match them with other symbols to produce correct replies. To outsiders, it appears the person understands Chinese, but in reality, they are only following rules. Searle argues that similarly, a computer may process language convincingly without genuine understanding or consciousness.

by Sampath Anson Fernando