Making Sense of the Senses

[This is Part 1 of a 3-part essay. Parts 2 and 3 will appear in a few days.]

Can you answer the following questions?

1) How many human senses are there? What are they?

2) How many planets do we have in our solar system, and what are their names?

3) Name all the continents in the world, and point out each one on a globe.

Before we discuss the answers to these questions, let’s first think about the questions themselves. What exactly are we doing when we ask questions such as these? Each of these questions is based on a classification system. So when we ask questions such as these, we are testing how well the responder has mastered the details of the classification system. With these 3 questions, we assume the existence of three distinct classes – senses, planets, and continents. We also assume that for any “thing” that we can name, we can categorize that thing as either being in the class or not in the class. For example, a turtle is not in any of these three classes – it is not a sense, a planet, or a continent. Jupiter fits into one of our three classes, because it is a planet – but it is not a sense or a continent. Africa is a member of another class, because it is continent – but it is not a sense or a planet. And taste is a sense – but it is not a planet or a continent.

As we will see in later essays, there is more than one way to structure a classification system. However, the three classification systems above – senses, planets, and continents – share a similar structure. In each case, the class contains fewer than 10 items, so we can easily memorize the entire list of “things” in the class. Furthermore, we can easily discuss the class in isolation from all other classes. For example, when we first teach children about planets, we don’t usually discuss asteroids and comets – that comes later in their education. When we first teach children about continents, we don’t expect them to simultaneously learn the locations of all the countries in the world. But later on, when the kids start to learn the locations of various countries, the most important location identifier is the continent in which the country is located.

Did you find that these three questions all look familiar? Basic classification models such as these are a fundamental part of what we teach our kids in school. Once our kids start kindergarten, classification models are a key component of the curriculum from that point on. Furthermore, for each of the above questions – and for other questions of the same type – we expect our students to provide the one acceptable “right” answer. And yet in each case, as we’ll see shortly, it is debatable as to whether the “right” answer is the only reasonable answer.

Human Senses

Question: How many human senses are there? What are they?

The standard answer, of course, is that there are 5 human senses – sight, hearing, taste, smell, and touch. This answer is so embedded in our culture that we have a common phrase – “sixth sense” – that refers to a person’s ability to perceive things that most other people cannot perceive. And yet the traditional answer of 5 human senses is not the only reasonable answer to the question. Imagine that there is someone — say, a visitor from another planet – who is unfamiliar with the usual answer to the question. Now imagine that this visitor is given the task of itemizing all of the human senses. How many senses would be included in the visitor’s list, and what would those senses be?

Because the visitor is unfamiliar with the traditional answer, he must use a set of criteria – rules of thumb – to sort through the possible answers. The criteria are derived from the definition of the class. So the first step is to define the word “sense”. What exactly is a “human sense”?

It turns out that the wording you use to define the phrase “human sense” can have a big impact on how many “things” can be included in the class. However, I would define “human sense” as “the ability of a human body to detect external matter, external forces, or external sources of energy”. This detection is enabled by “sense organs” within the body. Therefore a more complete definition would be “the ability of a human body to detect external matter, external forces, or external sources of energy, using sense organs within the body”.

Based on the above definition, anything that we would consider to be a human sense should meet the following criteria:

1) The sense enables us to detect external matter (that is, matter that is not part of our body) or an external force or source of energy.

2) It is possible to explain the nature of the matter, force, or energy that is detected by this specific sense.

3) It is possible to identify the specific part of the body that serves as the sense organ for this sense.

4) It is possible to explain how the sense organ detects the external matter, force, or energy.

The five traditional senses all meet the above criteria, as summarized below:

Sight – detects light. The sense organs are the eyes. Receptors in the retina of the eye respond to certain wavelengths of light. There are 4 types of receptors in the retina, each responding to a specific range of wavelengths. This allows us to distinguish among the wavelengths, and we perceive the various blends of wavelengths as different colors. The optic nerve conveys the data from the eyes to the brain. The brain processes the data to construct our sense of vision.

Hearing – detects sound waves in the air. The sense organs are the ears. Sound waves in the air trigger vibrations in the bones of the inner ear, which are detected by “hair cells” that send the data to the brain via the auditory nerve. The data includes information such as the mix of frequencies (pitches) in the sound and the loudness of the sound.

Taste – detects certain kinds of molecules that come in contact with the tongue and other tissues inside the mouth. The sense organ is therefore the mouth. The receptors are located in “taste buds” scattered along the surface of the tongue and on certain other tissues such as the soft palate. There are at least 5 kinds of taste buds, capable of detecting sweet, salty, sour, bitter, and savory molecular compounds. As with the other senses, the data is transmitted to the brain via the nervous system, and the brain processes the data to result in our perception of taste. (However, when we eat food, the brain blends taste data and smell data to create our perception of taste.)

Smell – detects certain kinds of airborne molecules that enter the nose. The sense organ is therefore the nose. The receptors are located in the olfactory bulb within the nasal cavity. There are apparently hundreds of different kinds of olfactory receptors, providing us with the ability to detect and distinguish a wide range of odors. The data is conveyed to the brain via the olfactory nerve, and the brain processes the data to create our perception of smell.

Touch – detects objects which come in contact with our skin. The sense organ is therefore the skin. The receptors are located beneath the skin, and detect pressure upon the skin. The wide distribution of the receptors allows us to deduce various properties of the object touching us, such as its texture and size. As with the other senses, the data is transmitted to the brain via the nervous system, and the brain processes the data to produce our sense of touch. In addition to the pressure receptors, there are several other kinds of receptors in the skin, including hot, cold, and pain. However, these various sensations are traditionally lumped into the sense of touch.

Okay, so we know that our visitor from another planet will include at least 5 human senses in his list. But will he find any other human senses to add to his list?

The visitor – let’s call him Marvin – does indeed catalog some additional senses. The next sense on his list is the sense of balance, and this is what Marvin has to say about it:

Balance – detects the direction of the force of gravity, allowing humans to determine whether their bodies are exactly vertical or slightly tilted. Human have long bodies, yet they walk on just two feet, balancing their long bodies perpendicular to the ground. Their sense of balance is extremely helpful when performing such activities as standing or walking, allowing them to maintain their balance and not fall over. The sense organ is the located in the inner ear (the semicircular canals), but is quite distinct from the sense of hearing. Fluid in the inner ear reacts to gravity, allowing the organ to detect the direction of the gravitation force. The fluid also reacts to acceleration forces, such as movements of the head. This information is conveyed to the brain via the vestibular nerve.

Marvin also draws a distinction between the sense of touch and the sense of hot and cold. Therefore, in Martin’s list we see the following seventh sense:

Hot and Cold – detects heat and coldness. The sense organ is the skin. Receptors located in the skin react to increased or decreased heat energy. However, these are not the same receptors responsible for the sense of touch. The change in heat energy may be due to conduction – either from a hot or cold object touching the skin, or from hot or cold air or water touching the skin – or the change in heat energy may be due to a net gain or loss by means of thermal radiation, such as feeling the warmth of sunlight or the heat from a fire. As with the other senses, the data is transmitted to the brain via the nervous system, and the brain processes the data to produce the sense of hot or cold.

The upshot is that Marvin’s list itemizes 7 human senses:

1) Sight (vision)
2) Hearing
3) Taste
4) Smell
5) Touch
6) Balance
7) Hot and cold

On the other hand, some people who study human senses say that there are 9 human senses. The other two senses, not on Marvin’s list, are the sense of pain, and the kinesthetic sense. (The kinesthetic sense, also called proprioception, is the ability to sense the position in space of various parts of the body.) However, Marvin did not include these latter two senses in his list, because in his opinion they do not involve the detection of external objects, forces, or sources of energy.

So now we have 3 different lists of human senses – one that lists 5 senses, one that lists 7 senses, and one that lists 9 senses. Can we really say that one of these models is “correct” and that the other two are “wrong”? In fact, each of these three models has a certain value. So what should we teach our kids? Should we teach them all 3 models?

Personally, if I could choose which model to use for instruction in schools, I would choose the model of 7 human senses. The 9-sense model would be my second choice. The 5-sense model is the weakest, because it limits our thinking too much. Any yet, teaching any of these models to our kids – even the 5-sense model – provides a lot of value. Any one of these models provides a framework upon which our kids can attach knowledge that they acquire. Therefore any of these models facilitates additional learning.

I would definitely not recommend that we present all 3 models on an equal footing to our kids. We should present one model as the standard, especially to younger children. However, we should not hide the fact that other models exist. As the kids get a little bit older, and have already become quite familiar with the standard model, it can be quite beneficial to point out that other non-standard models exist. Discussing these alternative models with our kids can significantly stretch their imaginations, helping them to see things in a different way. This is good training for being able to “think outside the box” as they grow older.

As a side note, we might ask the question, “If the 5-sense model of human senses is the weakest model, then how did the model originate, and how did we get so attached to it?” In my opinion, the 5-sense model is not really a model of the senses, but a model of the externally visible sense organs. There are 5 such organs – eyes, ears, mouth, nose, and skin. I believe that the focus on externally visible sense organs is the reason for including only 5 senses in the traditional model – in a one-to-one correspondence with the 5 organs. We remain attached to this model, despite its weaknesses, because it is so deeply rooted in our culture. If, on the other hand, a science educator were to be given the task of creating a better model for use in the primary school curriculum, the educator would probably choose a model that includes more than 5 senses – perhaps 7 or 9.

Now, suppose that we were to extend the model to include not just human senses, but the senses of all vertebrate animals. Would this add any more senses to our list, or would the list remain unchanged?

The one strong candidate to add to our list is a magnetic sense of direction. Some animals, such as certain birds, have the ability to detect magnetic fields – in particular, the magnetic field of the earth. There is actually an organ in the animal’s body that detects the orientation of the earth’s magnetic field relative to the animal’s body – something like a magnetic compass. Animals that possess this organ have a built-in sense of direction that is not dependent upon landmarks or the position of the sun in the sky. Such an organ is especially useful to birds who migrate long distances over the ocean. Therefore, if our model is expanded to cover all vertebrate animals, then the 7-sense model should be expanded to include at least 8 senses.

As we look through the world of animals – and now let’s include invertebrates as well as vertebrates – we can see examples of senses that appear to be different than human senses. We see snakes sensing the world by flicking out their forked tongues. We see bats navigating by means of echo location. And we see insects using their antennae to gather information about the world around them. But upon closer examination, we see that all of these senses share certain characteristics with human senses. If we define the sense of the smell as the ability to detect various kinds of molecules in the air, then it doesn’t matter whether the animal uses a nose or some other organ to do so. If we define the sense of taste as the ability to detect various kinds of non-airborne molecules that come in direct contact with the sense organ, then it doesn’t matter whether the animal uses its mouth or some other organ to do so. If we define the sense of hot and cold as the ability to detect heat gain or loss – either by means of conduction or thermal radiation – then it doesn’t matter whether the animal uses its skin or some other organ to do so. Even the bat’s ability to navigate by echo location can be described as an extension to the sense of hearing.

So that concludes our examination into the first class on our list – human senses. What about the next class on our list – planets?

[To be continued... Part 2 of the essay will appear in a few days.]

Copyright © 2010 by R. Philip Bouchard. All rights reserved.

Posted in Book Excerpt - Draft Copy, Natural Sciences | Leave a comment

Rethinking the English Alphabet – Part 5

[ Note: Parts 1 through 4 appeared in earlier installments of my blog. ]

Part 5 – The Vowel Sounds

In parts 1 through 4 of this essay, we have been considering how we might (theoretically) rethink the English alphabet to create a much stronger correspondence between the letters and the sounds that they represent. But up to now we have only considered the consonant sounds. There are several reasons that the vowel sounds present a much more difficult situation:

1) We have only 5 letters in the Latin alphabet for representing vowels (a, e, i, o, u) – but by anyone’s count there are more than 5 vowel sounds in English.

2) The vowel sounds in English are far more variable than the consonant sounds. Among all the various dialects of English spoken around the world, most of the consonant sounds are pronounced in a consistent fashion. But the pronunciation of the vowels can differ considerably from one dialect to another. Indeed, it is the pronunciation of the vowels, perhaps more than any other factor, which distinguishes one English speaker’s “accent” from another.

3) The vowel sounds in English not only vary from place to place, but also they have gone through historical changes. The most notable evolutionary change in the vowel sounds was the “Great Vowel Shift” from 1450 to 1750, during which most of the vowel sounds changed significantly. Prior to 1450, the sounds represented by the vowel letters were essentially the same as in Latin or Italian. After the vowel sounds all shifted, we continued to use the old spelling of the words. For example, we used to pronounce the word “feet” like we now pronounce “fate”, the word “wipe” like we now pronounce “weep”, the word “boot” like we now pronounce “boat”, and the word “date” like we now pronounce “dot”. But as we adopted new words from other languages, where the letters represented different sounds, we also preserved the spelling of these words, despite the inconsistency. For example, the spelling and pronunciation of the word “machine” is due to its French origin. We’ve kept the Italian spelling and pronunciation of “lasagna”. And we’ve kept the Portuguese spelling of “piranha”, although we’ve changed the pronunciation a bit.

4) In English, unlike many other languages, we don’t freeze the position of our mouths during the pronunciation of most vowels. Instead, we change the shape of our mouths while pronouncing the vowel, so that the sound at the end of the vowel is different than at the beginning. As a result, most of our vowel sounds are “glides” or “diphthongs” instead of pure sounds. Even linguists don’t always agree as to which vowel sounds should be considered as glides, and which should be considered as diphthongs. A diphthong consists of two vowel sounds in quick succession, where both sounds have a similar strength. A glide also involves two vowel sounds in quick succession, but the glided vowel is weaker than the main vowel. Therefore the combination is not really two complete vowels, but one vowel with an attached glide. The distinction is subtle!

5) In English, we often “reduce” the vowel sound in certain unaccented syllables. (However, we don’t do this in all unaccented syllables.) The reduced vowel sound is generally pronounced in one of two ways – similar to the vowel in “but” or to the vowel in “bit” – regardless of what letter is used to represent the vowel. Such a reduced vowel sound is sometimes called a “schwa” in the phonics world, represented by the symbol “ə”. (However, in IPA the “ə” symbol is more narrowly defined, applying only to the sound in “but” and not the sound in “bit”.)

Now back to the question – how many vowel sounds are there in English? The following 16 one-syllable words could all be considered as having different vowel sounds. However, some of the words contain diphthongs (two consecutive vowel sounds), and thus can be eliminated from our count – provided that we can agree as to which sounds are diphthongs and which are not:

bat
bait
bet
beet
bit
bite
pot
boat
boot
cute
book
but
pout
void
pert
caught

To make headway on this question, we should start by focusing on a single dialect of English, so let’s start with standard American English, also called broadcast American English. This is the version of American English that most television broadcasters use, which in turn is largely based on the English spoken in the American Midwest, minus any distinctive regional features.

I’ll start by presenting the model that I was taught in grades 1 to 3 during the “phonics instruction” of my early elementary school years. Although this model is far from perfect, it provides a very good starting point for discussion. According to the model I was taught, English contains 12 vowel sounds, along with two diphthongs. The 12 vowel sounds in this model, each of which has a unique “phonics” symbol, are:

ā (called “long A”) as in “bait”
ă (called “short A”) as in “bat”
ē (called “long E”) as in “beet”
ĕ (called “short E”) as in “bet”
ī (called “long I”) as in “bite”
ĭ (called “short I”) as in “bit”
ō (called “long O”) as in “boat”
ŏ (called “short O”) as in “pot”
ū (called “long U”) as in “cute”
ŭ (called “short U”) as in “but”
ōo (called “long OO”) as in “boot”
ŏo (called “short OO”) as in “book”

In addition to the twelve basic vowel sounds, this model includes two diphthongs – “oi”/”oy” and “ou”/”ow”. Each of these two diphthongs is considered to be a combination of two basic vowel sounds, spoken in quick succession.

oi/oy as in “toy” or “noise”
ou/ow as in “cow” or “house”

(My phonics instruction also included a sound called “au”/”aw” as in “dawn” or “caught”, but this sound is not a diphthong, just a digraph. In fact, to my ear this was not a distinct vowel sound at all, so I simply ignored it.)

As a kid, I loved this aspect of phonics instruction, because it provided me with a clear model and a clear set of terms for categorizing and discussing the vowel sounds. However, I was always annoyed by the “long U” concept, which to me was not a vowel, but a consonant-vowel combination. I strongly felt that “long U” should be discarded, because any sound represented by “ū” could be more properly represented by “yōo” – and I expressed this opinion to my third grade teacher, who was polite but noncommittal to my argument. (Many dictionaries use a pronunciation guide that is nearly identical to the phonics system I learned – and in these dictionaries the pronunciation of “cute” is typically listed as “kyoot” – so you could say that the dictionary supports my third-grade argument.)

As a kid I perceived a second problem in the phonics system I was taught – and that had to do with the “r-controlled vowel sounds”. We were taught that there are 3 such sounds – “er”, “ar”, and “or”. (The digraphs “ir” and “ur” don’t count, because they are pronounced the same as “er”.) However, to me the “er” sound seemed to be a single pure sound – not a vowel followed by a consonant. My mouth does not move at any point while making the “er” sound – so where does the vowel end and the consonant begin? My conclusion was that “er” ought to elevated to the status of a distinct vowel sound – and that “ar” and “or” were therefore diphthongs – each involving another basic vowel sound following by the “er” sound. Needless to say, my third grade teacher was not interested in changing the phonics curriculum to align with my suggestions, although she was polite about it. Anyway, I learned as an adult that the r-controlled vowels pose a special problem for phonemic categorization, because in many English dialects, including most of those spoken outside the US, all trace of the “r” sound is eliminated. However, some linguists agree that “er”, when pronounced as in standard American English, can indeed be considered a pure vowel sound, and one of the IPA symbols used to represent this sound is “ɚ”.

By the way, the use of the terms “long” and “short” to categorize the English vowel sounds is somewhat misleading. In some languages, including German and Japanese, there is a genuine distinction between long vowels and short vowels. In these languages you can have two distinct vowels that are pronounced in exactly the same way, the only distinction being that the long vowel is pronounced for a longer period of time than the short vowel. This concept does not exist in English, where each of the “long” vowels has a distinctly different sound than the corresponding “short” vowel. However, in standard American English, the long vowels tend to be complex, involving glides or diphthongs, while the short vowels tend to be more pure, with no glide or only a small glide. This rule of thumb breaks down when you consider other dialects of English. In the southeastern US, where I grew up, some people speak dialects of English where many of the “short” vowels are pronounced with a complex sequence of sounds.

(Side note: When I was a young adult teaching English as foreign language in South America, the head of instruction at the institute where I taught was a native of Germany. Her biggest pet peeve regarding the American instructors was the tendency to refer to the English vowel sounds as “long” or “short”. She absolutely could not tolerate this, and would launch into a furious rant if she heard any of us using these terms.)

My childhood mental model of 11 distinct vowel sounds (having eliminated “long U”) was challenged when, as an adult, I began work on a computer-based project for phonics instruction in elementary schools. At this time I realized that about half of my colleagues made a phonemic distinction between pairs such as “cot” and “caught”, which sounded identical to me. At last I understood why most dictionaries also made such a distinction in their pronunciation guides. (For example, the American Heritage dictionary uses “ŏ” for one of these sounds and “ô” for the other.) I also came to realize that there were certain words that were not members of such minimal pairs, and yet which could be categorized as having either the same vowel sound as “cot” or the same vowel sound as “caught” – and I was amazed to learn that these colleagues of mine could accurately predict which symbol (“ŏ” or “ô”) the dictionary would assign the vowel in each word. So for these people, there were 12 distinct vowel sounds in English, compared to my 11.

ô as in “caught”

However, we can reduce the count of vowels, and thus the number of distinct symbols needed, if we consider some of these 12 vowel sounds to be diphthongs – that is, a combination of two simpler vowels. Linguists appear to be in universal agreement that the “ī” sound is actually a diphthong – two consecutive vowel sounds – which in the phonics notation would be equivalent to ŏ+ē or ŏ+ĭ. Linguists are divided as to whether “ā” is a glide or a diphthong, but the mouth dramatically changes shape during the pronunciation of the vowel – so to my ears it is clearly a diphthong, equivalent to ĕ+ē or ĕ+ĭ. Likewise, the mouth dramatically changes shape during the pronunciation of “ō”, but if this is a diphthong, then what simpler vowel constitutes the first part? The best candidate in American English is probably “ô”, the sound that some people use for pronouncing “caught”. So “ō” can be considered as equivalent to ô+ ōo or ô+ŏo. (In the UK, “ŭ” may be a better candidate for the first vowel in this diphthong.)

This brings us down to 9 distinct vowel sounds, with the resulting need for 9 distinct vowel symbols. However, both ē and ōo are somewhat glided – the mouth closes a bit during the pronunciation of the vowel. We could use this fact as an argument for also treating these two vowel sounds as diphthongs, thereby bringing the number of distinct vowel sounds down to 7 – if we can figure out an appropriate set of symbols for the two parts of the vowel sound. (We’ll return to this question in a moment.)

As we did with the consonants, we will create 3 versions of PEA (Phonemic English Alphabet) for the vowels. In PEA-1 we will minimize the number of additional letters by assuming that there are only 7 basic vowel sounds in English, and that the remaining vowel sounds can be treated as diphthongs. In PEA-2 we will assume that there are 9 basic vowel sounds in English, and that the rest can be treated as diphthongs. Thus the alphabet for PEA-2 will have more vowel letters than PEA-1 does. In PEA-3 we will limit ourselves to the 26 traditional English letters, which means that we have to assign digraphs to some of the basic vowel sounds, just as we did with certain consonants in PEA-3.

This leaves us with several issues to work out for PEA-1 and PEA-2:

1) Assuming that we want each letter in the alphabet to consistently represent a specific sound, then which five sounds should we assign the 5 traditional vowel symbols – a, e, i, o, and u?

2) Assuming that we treat all but 7 or 9 of the vowel sounds as diphthongs, then which of the basic vowel sounds will get assigned new symbols other than a,e,i,o,u – and what should these additional symbols be?

3) Assuming that all of the remaining vowel sounds can be treated as diphthongs, then what two component sounds constitute each diphthong? In other words, what pair of letters should we use for the diphthong?

The first question may be the easiest. For PEA-1 and PEA-2, I recommend that the five traditional vowel symbols – a,e,i,o,u – should represent nearly the same sounds that they represent in Latin or Spanish (just as they did in English prior to 1450). Furthermore, the symbols should represent vowels that are essentially “pure”, with little or no glide in the vowel. Using this standard, the pronunciation rules for written English will be much more consistent with other languages that use the Roman alphabet – which in turn will make it easier for non-English speakers to learn English, and for English speakers to learn other languages. Therefore the five traditional vowel symbols can represent the following phonemes:

“a” = “ŏ” or “ä” (IPA /ɑ/ or /ɒ/) as in bark, dart, or yard, or as represented by the “o” in hot or rod (see Note 1)
“e” = “ĕ” (IPA /ɛ/) as in bed, desk, or help
“i” = “ĭ” (IPA /ɪ/) as in bid, grit, or hilt
“o” = “ô” (IPA /ɔ/) as in born, lord, or tort (see Note 2)
“u” = “ŏo” (IPA /ʊ/) as in put, or as represented by the “oo” in book

Note 1: Some people may consider this to be two distinct vowel phonemes, but in standard American English this is generally a single phoneme.

Note 2: For people who distinguish between the vowels in “cot” and “caught”, this is also the vowel sound in “caught”.

The two remaining “pure” vowels are those corresponding to the phonics symbols “ă” (as in “cat”) and “ŭ” (as in “bug”). In IPA these sounds can be represented as /æ/ and /ə/, respectively. We could use either the phonics symbols or the IPA symbols – but I recommend using the two phonics symbols, because I think that they will be less confusing to people.

Now let’s look at the vowel sounds that we will consider to be diphthongs or glides, and therefore will represent using digraphs:

ā as in “bait” (diphthong)
ē as in “beet” (glide)
ī as in “bite” (diphthong)
ō as in “boat” (diphthong)
ōo as in “boot” (glide)
ou/ow as in “house” or “howl” (diphthong)
oi/oy as in “boy” or “void” (diphthong)

There are several ways we could approach this – but here is one simplifying assumption. In every glide or diphthong in standard American English, the mouth moves from a more open position to a more closed position. In fact, there are only two possible ending positions, roughly corresponding to the vowel sounds that we have assigned to the letters “i” and “u”. However, these ending positions also correspond to the semi-vowels “y” and “w” – and this correspondence feels more accurate. Therefore we can write all seven of the above sounds as a digraph in which the first letter is one of our 7 basic vowels, and the second letter is either “y” or “w”. As a result, the above seven sounds will be written in PEA as follows:

ā as in “bait” is the diphthong e+y, and therefore will be written as “ey”, as in “grey”
ē as in “beet” is the glide i+y, and therefore will be written as “iy” (PEA-1 only)
ī as in “bite” is the diphthong a+y, and therefore will be written as “ay”
ō as in “boat” is the diphthong o+w, and therefore will be written as “ow”, as in “blow”
ōo as in “boot” is the glide u+w, and therefore will be written as “uw” (PEA-1 only)
ou/ow as in “house” is the diphthong a+w, and therefore will be written as “aw”
oi/oy as in “boy” is the diphthong o+y, and therefore will be written as “oy”

Note that these are not arbitrary assignments. In each case, the first letter of the digraph corresponds to the “simple vowel” that is most similar to the beginning sound of the diphthong or glide. The second letter indicates which of the two semi-closed positions terminates the diphthong or glide.

Now let’s look at the three “r-controlled” vowel sounds – “er”, “ar”, and “or”. These pose a special challenge. In standard American English, the “r” in these combinations is clearly pronounced. But in some dialects of American English (such as Boston, and the coastal plain of the Deep South), the “r” is not pronounced. (Linguists use the terms “rhotic” and “non-rhotic” to make this distinction. Most dialects of English spoken outside of North America are non-rhotic.) To my ear, in standard American English “er” sounds like a unique, pure vowel, while “ar” and “or” sound like diphthongs in which the second vowel sound is “er”. However, one of our rules for PEA-1 is to minimize the number of symbols in the alphabet, while still maintaining a 1-to-1 correspondence between sounds and symbols. Therefore, for PEA-1 we will follow convention and consider the “r” in these r-controlled vowel sounds to be a normal consonant. The sounds “ar” and “or” will continue to be written as “ar” and “or”. In rhotic dialects such as standard American English, I find it hard to match the vowel sound in “er” to any of the other vowels already in the list – the “er” vowel sound appears unique to me. But in non-rhotic dialects, the sound I hear people speak sounds to me like the “u” in “but”, to which we have assigned the symbol “ŭ”. When people speaking such dialects say the word “butter”, it sounds to me like “bŭtŭ”. Therefore in PEA-1 we will assume that the vowel sound in “er” can be represented as “ŭ”, and we will use “ŭr” as the spelling for the “er” sound.

For PEA-2, when we discussed the consonants, we relaxed the rule to minimize the number of new symbols. The same concept will apply to the vowels. In PEA-1 we treated two vowels with minor glides as diphthongs:

ē as in “beet” will be written as “iy” (PEA-1 only)
ōo as in “boot” will be written as “uw” (PEA-1 only)

But for PEA-2 we’ll elevate these two sounds to the status of distinct vowels, each with its own unique symbol:

ē as in “beet” will be written as ï (PEA-2 only)
ōo as in “boot” will be written as ü (PEA-2 only)

The concept here is that the two dots indicate a glide at the end of vowel sound, which distinguishes the vowel from the similar vowel written without the two dots. So “ï” sounds a lot like “i”, except for the glide, and “ü” sounds a lot like “u”, except for the glide.

In PEA-2 we’ll also introduce a special variation of the “r” symbol – ř – to use with r-controlled vowels. This will allow us to treat the “er” sound as a unique, pure vowel sound, written as “ř”. In contrast, the “ar” and “or” sounds are considered to be diphthongs in PEA-2, and are written as “ař” and “oř”. This notation also works reasonably well for the non-rhotic dialects, because it clearly distinguishes between the consonant “r” (which is always pronounced) and the vowel “ř” which is either not pronounced or is reduced to something like “ŭ”.

So now let’s compare the spelling of several words in traditional English spelling, PEA-1, PEA-2, and two distinct interpretations of IPA:

Traditional Spelling IPA (1) IPA (2) PEA-1 Spelling PEA-2 Spelling
bat /bæt/ /bæt/ băt băt
bait /bet/ /bejt/ beyt beyt
bet /bɛt/ /bet/ bet bet
beet /bit/ /bijt/ biyt bït
bit /bɪt/ /bit/ bit bit
bite /bɑɪt/ /bɑjt/ bayt bayt
cot /kɑt/ /kɑt/ kat kat
caught /kɔt/ /kot/ kot kot
boat /bot/ /bowt/ bowt bowt
cut /kʌt/ /kət/ kŭt kŭt
cute /kyut/ /kyuwt/ kyuwt kyüt
book /bʊk/ /buk/ buk buk
boot /but/ /buwt/ buwt büt
pout /pɑut/ /pɑwt/ pawt pawt
void /vɔɪd/ /vojd/ voyd voyd
pert /pɚt/ /pərt/ pŭrt přt
part /pɑɚt/ /pɑrt/ part pařt
port /pɔɚt/ /port/ port pořt

Now, using either PEA-1 or PEA-2, we have a system for spelling English words that can provide a 1-to-1 correspondence between the sounds of the language and the symbols used to represent those sounds – in other words, a truly phonemic spelling system. However, because English has more than 26 consonant and vowel sounds, both PEA-1 and PEA-2 contain more than 26 letters. In PEA-1 we used a very conservative count of the unique sounds (phonemes), and we ended up with the following 29 letters in the alphabet:

a ă b c d e f g h i j k l m n o p q r s t ð u ŭ v w x y z

In PEA-2 we used a slightly more liberal approach for counting the number of unique sounds in the English language, and we ended up with the following 32 letters in the alphabet:

a ă b c č d e f g h i ï j k l m n o p r ř s š t u ŭ v w x y z ž

That leaves us with the final challenge. In PEA-3 we must restrict ourselves to the 26 traditional letters. Therefore we must use digraphs for some of the phonemes – where two letters written in sequence represent a single unique sound. (An example from the consonant phonemes is “sh”.) In PEA-1 we minimized the number of new symbols by adopting the convention that English has only 7 unique vowel sounds, and that all of the other vowel sounds are actually diphthongs. As a result we only had to introduce two new symbols for the vowels. If we adopt the same approach for PEA-3, where we use digraphs instead of new symbols, then we will have to employ at least two vowel digraphs.

A key rule of thumb for PEA-3, as we saw with the consonants, is to choose traditional digraphs that are readily recognized by current English speakers – especially if the digraphs help to make PEA-3 more consistent with the traditional spellings of words. The most obvious example is to use “oo” to represent the vowel sound in such words as “boot” and “noon”. (Note, however, that “book” has a different vowel sound, even though it too is traditionally spelled with “oo”.) So it might make sense to replace the “uw” diphthong in PEA-1 with the digraph “oo” in PEA-3. However, this does not actually eliminate either of the two new vowel symbols that we introduced in PEA-1, which are ă and ŭ.

The digraph “ah” is readily recognized and understood by English speakers, even though it is seldom used in the traditional spelling of words. This digraph corresponds to the symbol “a” in PEA-1. If we employ this digraph in PEA-3, then this frees up the symbol “a”, which we can then use to represent the vowel sound in the word “bat”, thereby eliminating the symbol “ă”. However, in this approach, the diphthong “ay” now has to be spelled “ahy”, and the r-controlled vowel “ar” now has to be spelled “ahr”. (On the other hand, the diphthong “aw” can be left alone, because it still makes sense even with the new vowel sound assigned to the symbol “a”.)

The digraph “uh” is also fairly well recognized by English speakers – although it is almost never used in the traditional spelling of words. If we replace the “ŭ” symbol in PEA-1 with the digraph “uh”, then the pronunciation of the digraph should be fairly clear. However, this is an extremely common vowel sound in English. In contrast, the vowel sound in “book”, represented by “u” in PEA-1, is relatively rare. So I suggest swapping the spelling convention for these two sounds. In PEA-3, the vowel sound in the word “but” will be represented by “u”, while the vowel sound in the word “book” will be represented by the digraph “uh”. This change will help to make the spelling under PEA-3 a bit more similar to traditional spelling. It also allows us to use “ur” instead of “uhr” for the r-controlled vowel. However, under this approach, it no longer makes sense to use “uw” to represent the vowel sound in “boot”, so we’ll use the traditional “oo” digraph instead.

Traditional Spelling IPA (1) IPA (2) PEA-1 Spelling PEA-2 Spelling PEA-3 Spelling
bat /bæt/ /bæt/ băt băt bat
bait /bet/ /bejt/ beyt beyt beyt
bet /bɛt/ /bet/ bet bet bet
beet /bit/ /bijt/ biyt bït biyt
bit /bɪt/ /bit/ bit bit bit
bite /bɑɪt/ /bɑjt/ bayt bayt bahyt
cot /kɑt/ /kɑt/ kat kat kaht
caught /kɔt/ /kot/ kot kot kot
boat /bot/ /bowt/ bowt bowt bowt
cut /kʌt/ /kət/ kŭt kŭt kut
cute /kyut/ /kyuwt/ kyuwt kyüt kyoot
book /bʊk/ /buk/ buk buk buhk
boot /but/ /buwt/ buwt büt boot
pout /pɑut/ /pɑwt/ pawt pawt pawt
void /vɔɪd/ /vojd/ voyd voyd voyd
pert /pɚt/ /pərt/ pŭrt přt purt
part /pɑɚt/ /pɑrt/ part pařt pahrt
port /pɔɚt/ /port/ port pořt port

In each of the three versions of PEA, any given phoneme is always written in a consistent fashion, using a unique letter symbol, or a unique pair of letter symbols. The table below lists all of the consonant phonemes of English, along with the corresponding spelling for the phoneme:

Phoneme (IPA) PEA-1 Spelling PEA-2 Spelling PEA-3 Spelling
/p/ p p p
/b/ b b b
/t/ t t t
/d/ d d d
/k/ k k k
/ɡ/ g g g
/tʃ/ tx č ch
/dʒ/ dj j j
/f/ f f f
/v/ v v v
/θ/ c c c
/ð/ ð x th
/s/ s s s
/z/ z z z
/ʃ/ x š sh
/ʒ/ j ž zh
/x/ x
/w/ w w w
/l/ l l l
/ɹ/ r r r
/h/ h h h
/m/ m m m
/n/ n n n
/ŋ/ q q ng
/j/ y y y

The following table lists all of the vowel phonemes for standard American English, along with the corresponding spelling in each of the three versions of PEA.

IPA (1) IPA (2) Phonics PEA-1 PEA-2 PEA-3
/i/ /ij/ ē iy ï iy
/ɪ/ /i/ ĭ i i i
/e/ /ej/ ā ey ey ey
/ɛ/ /e/ ĕ e e e
/æ/ /æ/ ă ă ă a
/u/ /uw/ ōo uw ü oo
/ʊ/ /u/ ŏo u u uh
/o/ /ow/ ō ow ow ow
/ɔ/ /o/ ô o o o
/ɑ/ /ɑ/ ŏ a a ah
/ɚ/ /ər/ er ŭr ř ur
/ʌ/ /ə/ ŭ ŭ ŭ u
/ɑɪ/ /ɑj/ ī ay ay ahy

We are nearly finished now – our definition of the Phonemic English Alphabet is essentially complete. In the final installment of this essay, coming in a few days, we will transcribe a few sample texts from the traditional English spelling into PEA-1, PEA-2, and PEA-3. At the same time, we will address the final few questions, such as what punctuation conventions to use.

Copyright © 2010 by R. Philip Bouchard. All rights reserved.

Posted in Book Excerpt - Draft Copy, Languages & Linguistics | Leave a comment

Melbourne, Australia

[ Tomorrow I will have completed 3 weeks in Australia – one week each in Sydney, Melbourne, and Brisbane, which are Australia’s three most populous cities. Today’s blog entry is about Melbourne. ]

Melbourne is the southernmost of Australia’s major cities, and as a result it has the coolest climate. My visit occurred during the last week of October, equivalent to the last week of April in the northern hemisphere, and I found that most people in Melbourne were still wearing coats and jackets, especially in the late afternoon.

Melbourne, Australia

I was quite impressed with downtown Melbourne. The Central Business District (CBD) is laid out in a grid pattern along the north bank of the Yarra River. There is a nice mix of traditional and modern architecture, with quite a few skyscrapers.

Melbourne, Australia

The downtown streets are wide and lined with sycamore trees. Most of the people walking around downtown are dressed in nice clothes – but most of the clothing is black. The combination of the architecture, trolley lines, and dark clothing gives the downtown a rather formal or traditional appearance.

Melbourne, Australia

Public transportation is primarily based on trams (trolleys or streetcars) and trains, although out in the suburbs you also see buses. The entire central business district is crisscrossed by a network of tram lines, which then radiate out into the suburbs. The train lines form a loop around the central business district and then also radiate out into the suburbs. The tram lines generally run down the middle of the streets, while the train lines are grade separated from the automobile traffic.

Melbourne, Australia     Southern Cross Station - Melbourne

Beginning at the western edge of the central business district, the old Docklands have recently been redeveloped into trendy new real estate, with quite a bit of adventurous architecture.

Docklands - Melbourne, Australia

Perhaps what impressed me most about Melbourne is the abundance of green parkland within a few miles of downtown. And these parks aren’t small – they are huge. Many of these parks are located along the Yarra River or along other streams. This is turn has allowed the creation of an interconnected network of walking and bicycle trails.

Melbourne, Australia

The city of Melbourne strongly encourages cycling as an alternative form of transportation. At rush hour you’ll see a swarm of bicyclists making use of the bike routes. There are blue bicycle rental stands all over downtown, allowing anyone with a membership card to pick up a bicycle at one location and drop it off at another. It’s like the ZipCar concept, except for bicycles.

Melbourne, Australia     Bicycle Rental - Melbourne

Finally, I’ll end this blog entry with a note about socializing in the central business district. In most of downtown Melbourne, it appeared that most people were simply trying to get from one point to another. But in Federation Square, across from Flinders Street Station, people were content to sit and relax at any time of the day.

Federation Square - Melbourne

And at the southern edge of Federation Square, along a wall forming an embankment of the Yarra River, an incredible outdoor bar scene developed immediately after work hours – people clearly enjoying themselves!

Happy Hour - Melbourne, Australia

Copyright © 2010 by R. Philip Bouchard. All rights reserved.

Posted in Diary, Visiting Places | Leave a comment