Understanding scientific principles by using them to prove the structure of words

I am using this post on Real Spellers to share some spelling investigations and resources sparked by an on-line consultancy with teachers and students at the International Schools Riau in Sumatra, Indonesia. Iʼm leading a series of lessons via Skype with four Middle School classes there. We have just finished our second session working together in lessons which integrate structured word inquiry with their studies of the United Nations Millennium Goals.

I hope to expand the community of learners who can gain from, and contribute to these investigations by sharing some of this early work on Real Spellers. (Note: Embedded in the text are links to relevant teacher resources to download)

Structuring instruction to develop “sustainable” scientific inquiry knowledge!

One of my main goals is to help these teachers and students take the role of “word scientists” who have the knowledge and tools that hold them in good stead for investigating the spelling meaning connection of words long after our work together.

With that purpose in mind, I tried to use words from their unit on millennium goals to highlight scientific principles that guide any kind of scientific inquiry -- not just scientific analysis of word structure and meaning.

There are two related scientific principles I want to highlight that can be stated like this:

Scientific inquiry seeks:

  1. The deepest structures that account for the greatest number of cases;
  2. The most “elegant solution” -- that which requires the fewest “entities” or assumptions is best.

(So far I have only explicitly referenced the first with these classes.)

These principles go back to Occamʼs razor, which my Oxford defines as, “the principle (attributed to William of Occam) that in explaining a thing no more assumptions should be made than are necessary.”

I know these ideas sound fairly abstract out of context. My hope is that I can help clarify the utility of these ideas for teachers and students by making explicit reference to them as we use them to determine why one scientific analysis of the spelling of a word is better than another. In this way our understanding of scientific inquiry and our ability to investigate the spelling / meaning connections will grow.

As you read the investigations we have begun in these classes, consider how we are using these principles to come to come to conclusions of which hypothesis is the best description of how spelling works.

Our Investigations...

The audience I am working with already has a good background in the basics of English spelling from working with Real Spelling and WordWorks. One of the spelling concepts teachers requested I address had to do with double letters, so we started out with some practice making word sums and spelling out word structure with words on a sheet I prepared for the classes. You can find all of the documents I prepared for these classes at this link

(If you choose to try this challenge yourself, you might also find this document on announcing word structure useful. That document is part of a longer post at this link.)

It didnʼt take these students long to get comfortable with announcing word structure with double letters and dropped single silent <e>s. To analyze words from their list, I also had to ensure that they were comfortable with bound bases. It was quickly clear these classes could work well with all these concepts. With that background established, I set a task for our session a couple weeks after the first session.

The second page of the first document has a set of words I chose because of their association with UN Millennium goals. This topic made it easy for me to find words that would force students and teachers to think carefully about announcing double letters, bound bases, and other spelling features and conventions.

Student questions in the follow-up lesson

I was delighted with the thinking sparked by the task of analyzing and announcing the word structure of their vocabulary words. Students and teachers developed a number of great observations and hypotheses to test.

<-tion> or <-ion>

In one class, the words <imagination>, <education> and <globalization> sparked a debate about whether there is a <-tion> and an <-ion> suffix, or if there is only an <-ion> suffix. A team of two students from one class described their conclusion that there was no <-tion> suffix despite the fact that their dictionary listed one. They found no word that could be analyzed coherently with <-tion>. (It is plain that there is no <*imagina>, <*educa>, or <*globaliza>!)

With that evidence, they rightly concluded that <*-tion> does not act as a suffix. Until a word is presented with a coherent word sum using a this suffix, weʼll just have to disagree with their dict + ion + ary on that one! (Perhaps we should in + dict their dictionary for violence aganinst, or at least violation of morphological structure!)

The structure of <millennium>

(Note: Click here for the student sheet that grew from this investigation.)

Among the other great questions from one of the Grade 5-8 classes were some on the structure of the word <millennium>. Between our sessions, their teacher passed on questions from her students:

What is the base in the word millennium? Is it <mill>, <mille> or <mil>? Most students were in favor of <mille> but it was not decisive. We looked at <million>, <millimeter>and <milliliter> for exploration.

Since they were focused on the correct spelling of the base for ʻthousandʼ, I shared my analysis of the rest of the word sum that includes a bound base <enn> for ʻyear,ʼ a connecting vowel letter <-i-> and a Latin suffix <-um>. Combined with their hypotheses for the other bound base in this word, we were left with three possible word sums:

mill + enn + i + um → millennium

mille/ + enn + i + um → millennium

mil(l) + enn + i + um → millennium

I studied this question before our session and was fairly confident I knew the correct answer, but as I told the class, I still needed more evidence before I felt totally comfortable with my answer to this question.

We worked through the three hypotheses together. I was impressed that their hypotheses included reasonable applications of the suffixing conventions for doubling consonants and replacing single silent <e>s.

And it was through describing those suffixing conventions in conjunctions with their hypotheses that I recognized the evidence I needed to be confident in my conclusion. I always love it when I discover a tangible new fact for myself in the context of teaching a class! I'm always looking for opportunities for "inquiry-led teaching" but you never know when they are going to pop up. 

Iʼll let Real Spellers readers consider why the process of reviewing the suffixing conventions with students helped me see that if the word <millennium> uses a bound base <enn> for ʻyear,ʼ the initial bound base must be spelled <mill>.

Turning to the structure of <universal> and <development>

There were many excellent questions from the other two classes as well. I offered brief responses to a question about the structure of <development> and how it relates to <envelope>. This is another commonly occurring question, as the lack of an <-e> suffix(as far as I know!) prevents this word from fitting in a matrix with the base <velop>. This group was already on the right track with their etymological research.

They also wondered about the word <universal>. They asked how it relates in meaning to these words they found on the word searcher: university, universe, diverse, diversity, version, verses, verse, reversible, reverse. I acknowledged that it might seem that <universe> has nothing to do with a “verse of poetry”. We didnʼt have time to go into detail yet, but I provided the smallest of hints by encouraging the class to consider the expression “turn of phrase.” To see how these words share a commonly spelled base, readers will need to remember their suffixing patterns for the single silent <e> and vowel suffixes. (Find two types of suffixing checkers with those conventions here.)

Investigating the structure of <sustainability>

(Note: Click here for the student worksheet that grew from this investigation) 

We could have spent a session on any of these questions, but I decided to focus our time on a question about the structure of <sustainability>. This was a great opportunity to make sense of a suffix that regularly causes students and teachers difficulty. And in doing so, I hoped to do an investigation that made clear use of the scientific principles that I wanted to target.

To repeat, those principles are that scientific inquiry seeks the deepest structures that account for the greatest number of cases, and that it the best analysis avoids unnecessary assumptions or “entities.”

The class offered two tentative hypotheses of the morphological structure of this word:

<sustain+ab+il+ity> and <sustain + abil + ity>.

They also asked and if <abil> might be a version of the word <able> (or perhaps the suffix <-able>.

I pointed out that they were correct to analyze <sustain> as connected in structure and meaning to <sustainability>. It is true that there is an <-ity> suffix, but that we had to prove whether their could be an <-ity> suffix in this <sustainability>. I also shared that many people struggle with how to analyze the letter sequence <ability> (and <ibility>) at the end of words, so this was a good word to investigate carefully.

We used the word searcher to test their hypothesized suffixes <-ab>, <-il> and <abil>.To determine that any of these were possible suffixes, we would have to find words which met two criteria that define suffixes.

  1. Suffixes can be final.

By typing these letter combinations followed by a dollar sign in the Word Searcher, we could search through 60 000 words to find words with these final letter sequences.

  1. Suffixes are fixed to a base or stem.

This second point means that finding this letter sequence in final position is not sufficient. Any such word would have to be analyzable into a coherent word sum. For a clear example, the word <spring> has a final <ing>, but we can see that it does not use an <-ing> suffix as there is no such thing as a base or stem spelled <*spr>!

With our cirteria set, we could safely test which hypothesis held up  to the evidece -- that is which hypothesis identified the deepest structures that account for the greatest number of cases wihtout proposing “unnecessary entities.”

Testing for <abil>:

We found zero cases of a final <abil> in final position. Thus we could eliminate this hypothesis as a possibility.

Testing for <ab> and <il>:

We did find words with these letter sequences in final position, but none of them analyzed into a coherent word sum with this letter sequence as a suffix. For example, <stab> has a final <ab>, but since <st> doesnʼt make any sense, so this word could not be evidence of an <*-ab> suffix.

As scientists, we had no option but to conclude that there no evidence to support these hypothesized suffixes. But as scientists know, proving hypotheses false is not a failure, but a success. By eliminating reasonable, but false leads, we could direct out thinking to other possibilities.

I asked the class what we are left to conclude about the structure of <sustainability> if we determine that there is no <*-ab>, <*-il> or <*-abil> suffix?

Our conclusion was that since <sustain> was a word that made sense in structure and meaning within <sustainability>, the only coherent solution is that this word uses a suffix spelled <-ability>. People may sense that this is “too long” to be a suffix. But of course we can’t come to conclusions about what we think seems right or wrong. As word sientists, all we can do is follow the principles of scientific inquiry. Scientists must adhere to scientific principles whether they lead us to conclude that the Oxford English dictionary is in error when it cites <*-tion> as a suffix in <completion> or <relation>, or whether it leaves us to conclude that <-ability> is a suffix -- clunky or not!

With a little more work with etymology on-line and the word searcher, we were able to prove that <sustain> was in fact a stem, not a base. It is composed of prefix <sus-> and a bound base <tain>. We found words like <contain> and <retain> that share the structural element <tain> and the Latin root tenere for ʻto hold.ʼ

With that discovery, were were able to start a matrix with this bound base.

The next challenge

With all of that work established through these investigations, I would like to invite Real Speller readers to build on this foundation. I propose three questions in line with the learning goals I have for the students and teachers in Sumatra. You may, of course, have another point that interests you more.

  1. Explain how the bound base <tain> with itʼs Latin root tenere for ʻholdʼ informs the meaning of the word <sustainability> with reference to millennium goal “Environmental Sustainability”
  2. Consider the evidence for the following conclusions described above:
  1. There is no <*-tion> suffix.
  2. The word <sustainability> must use a suffix spelled <-ability>.

Explain how either of these conclusions draw on the scientific principles identified at the outset. To repeat yet again, those principles state that the best scientific solution identifies the deepest structures that account for the greatest number of cases, and doesnʼt create unnecessary “entities” or assumptions.

2) Some time ago I worked with a student in a class during my visit to Doha. we investigated these three word sums for the word <structured>:

struct + ured → structured

struct + ure + d → structured

struct + ure/ + ed → structured

Drawing on these same principles, which of these word sums provides the correct (the most elegant) analysis of this word?

AFTER you come to your own conclusion for this question, you might enjoy watching this video of a Grade 5 class at the American School of Doha where a student I was working with brilliantly sorted out this same principle with a word of a parallel structure.


I hope students and teachers around the world take advantage of the foundation of learning offered by the Grade 5-8 students at the International Schools Riau in Sumatra, Indonesia, and help push all our our learning farther by investigating these questions, or better yet, coming up with some of your own!