Real Spellers

English Makes Sense!

Hey all,

I had the pleasure of diving into words with my daughter’s Grade 4-5-6 class this morning and as usual some interesting questions popped up. I wanted to share those questions with Real Spellers to see if we might get some ideas from the wider community out there.

As I started to recount the story, I realized that these questions also provide a great opportunity to identify guiding principles of morphological problem-solving as well. I hope that the aside to identify the scientific principles that I try to use to guide my investigation of novel spelling questions provides the reader with useful guidance in their own investigations.

So let’s start!

One student asked me about the word <dramatic>. My knee-jerk reaction was to assume the base <drama>, but a quick word sum revealed a problem:

drama + tic → dramatic ???

I have no knowledge of, nor can I find any evidence of a <-tic> suffix. So what are my options?

Conveniently for me I encountered this student’s question the day after I worked though similar questions in a Skype session with a weekly “study group” of teachers that has formed at The Nueva School since my visit there in October. As a result of that Skype session, I had my own advice to the Nueva teachers fresh in mind as students here in Kingston peppered me with great questions about all sorts of words after I set them off in search of new affixes. I still benefit from explicitly reminding myself of how to respond scientifically to a hypothesized word sum that is tempting, but which has elements that I cannot prove.

Without evidence of a <-tic> a suffix, a scientific inquiry has to give up on the hypothesis that <drama> is the base of <dramatic> -- no matter how tempting it might be!

(How can I prove a suffix? Find at least two words that use the same letter string at the end of a base or stem for a parallel purpose.)

A scientific response to a word sum with unproven elements should be guided by the scientific principle of elegance which includes the idea that scientific inquiry seeks the deepest structures that account for the greatest number of cases.

From the scientific principle of elegance, I draw the following principle to guide any morphological analysis:

  • Never accept an analysis that goes deeper than can be proven.

What does all of this have to do with the my hypothesized word sum for <dramatic>?

If my first hypothesis that <dramatic> has the base <drama> does not stand up to analysis with a word sum, I have to fall back to the deepest structure for which I do have evidence -- in this case accepting <dramatic> as a base word. This analysis may or may not be a full analysis, but it does not analyze deeper than I can prove.

If further evidence presents itself of another way to analyze this word in the future, I can refine my analysis. Critically, however, I don't make the error of falsely analyzing a word in  way that violates morphological structure.

Types of morphological analysis errors

It is important to be aware of the two kinds of morphological analysis errors that we can make as scientists trying to ascertain the deepest morphological structure of a given word.

  1. Incomplete analysis: This kind of error does not pose a barrier to learning. If at a given moment my orthographic knowledge is not sufficient to fully analyze a word, an incomplete analysis provides a "holding position" which can always be refined in the event that new relevant information is identified to resolve a word sum more fully.
  2. False analysis: This kind of error can be a serious barrier to learning. By false analysis I mean analysis that violates a morphemic boundary (placing a plus sign of a word sum inside a morpheme). This kind of analysis is a barrier to learning because it degrades one’s ability to recognize the meaning cues spelling provides through morphological conventions.

I will return to the question of the structure of <dramatic> shortly, but it may be helpful to  understand these distinct error types by considering another word. Let’s consider these hypothesized word sums for the word <relation>.

A) rela + tion → relation

B) relate/ + ion → relation

Hypothesis A is a false analysis of the morphological structure of <relation>. What is my evidence for that conclusion? Firstly, I have yet to see any evidence of a letter sequence <tion> acting as a suffix (a single morpheme added to the end of a base or stem) in any word. Until such evidence is presented, there is no justification for labeling this as a suffix. Secondly, there is no evidence for a base spelled <rela>. The word sum in Hypothesis A violates the morphological structure of this word because it places a plus sign inside a morpheme, not between proven morphemes.

How does this false analysis hinder learning? If a learner accepts <*-tion> as a suffix, his or her ability to recognize the orthographic cues that link the word <relation> to words like <relate> or <relative> is degraded. Further, creating a false analysis like <*rela + tion → relation> builds a mental representation of a false suffix in the learner. Students who look for <*-tion> suffixes as a result of this instruction are hindered from accurate morphological awareness when encountering words like <question>, <action>, <traction> or any word ending in <tion>. Focusing on the <tion> in these words leaves non-morphemes <*ques>, <*ac> and <*trac> which are less likely to spark links to other morphologically related words.

Building mental representations of false structures generates an extraneous load to morphological problem-solving. Thus instructional principles should be designed to avoid the error of false analysis.

Hypothesis B is an incomplete analysis. It turns out that the full analysis of <relation> is <re + late/ + ion → relation>.

My Word Stems dictionary by John Kennedy provides evidence that <late> is a bound base from the Latin tollere, latus for ‘carry, lift, bring’. (It happens to be a homograph for the bound base <late> for “not on time”.).

The incomplete analysis of Hypothesis B does not hinder my ability to recognize the full analysis of this word  -- when I develop enough of the background knowledge to recognize <relate> can be further analyzed. Some orthographic concepts that help me analyze <relate> include knowledge about bound bases and homographs. Over time, encountering words like <collate>, <elate>, <translate> and familiarity with the prefixes <col->, <e-> and <trans-> may spark a new testable hypothesis about the structure of <relate>.

Notice, however, that a false analysis such as <*rela + tion> would have to be identified as false and then re-analyzed for words like <collate>, <elate> and <translate> to be effective cues for the further analysis of <relate>.

  • The principle of not accepting an analysis that is deeper than I can prove is a strategy which avoids false analysis errors, in favour of errors of incomplete analysis.

Back to <drama> and <dramatic>!

Applying this principle of not analyzing deeper than I can prove resulted in my telling the student who asked me about the word <dramatic> that this was a great question that I had not considered before. I could signal my own scientific principle that until I had evidence of a <-tic> suffix, I would have to treat <dramatic> as a base. However, I could also signal that I was very curious about this question and that I was going to have to do some more work before I was convinced that <dramatic> was really a base. I did tell her  that because of the clear meaning and spelling similarities, I suspected that at the very least these words shared a common root origin, but so far I couldn’t prove that they had a common morphology. I had no evidence that they shared a base because I couldn’t make a word sum work.

The distinction between morphological relatives and etymological relatives is such a crucial understanding for effective scientific inquiry into spelling-meaning connections. It is also one of the key concepts I addressed with the Nueva teachers the night before. This is a question that comes up so often that I have this page on this topic on my website that I recommend.

When I got home, I decided to take a quick look in my Oxford dictionary on my Mac to double check the origin of the words <dramatic> and <drama>.

Here’s what I get for <dramatic>:

ORIGIN late 16th cent.: via late Latin from Greek dramatikos, from drama, dramat- (see drama ).

Hmm.  This got me thinking about another possibility for a morphological word sum that had not occurred to me at first.

I now wanted to check another etymological resource. I went to Etymonline to see what I would find there:

1580s, from L.L. dramaticus, from Gk. dramatikos "pertaining to plays," from drama (gen. dramatos; see drama).

A new hypothesis of the structure of <dramatic>!

With this new etymological information, I now have a new hypothesis for a word sum for <dramatic>, but one which does not use a base spelled <drama>.

I’m not going to share that hypothesis here now, however. I’m curious to see if this evidence sparks some ideas from teachers/students out there.

A weakness in my own understanding is knowing how to interpret etymological information to help me prove the structure of a word sum. Knowing this is an issue for me, I am sure it is for many out there. I’d be very curious to see if my colleagues at the Nueva School (or any other orthographic laboratories of students and teachers/tutors out there) come up with a hypothesis similar to mine before I share my current hypothesis. If so,  I would love to work with that group to present a clear hypothesis statement and ask some of the Real Speller community members with more expertise in etymological concepts to help us.

Before I leave you all with this question, however, I do need to make sure that you have access to all the types of written morphemes out there. For that reason, I’m including this image of my “morphological tree” of terms from my teacher resource book.


Happy spelling!



Comments (0)

There are no comments posted here yet