Real Spellers

English Makes Sense!

We all know that this word is spelled correctly:proper.

In my mind, the base word is prop and the suffix-er is added. In looking at the Big Suffix Checker I follow this sequence: does the suffix begin with a vowel letter? (Yes), does the word taking the suffix end with a vowel letter? (No), does the word taking the suffix end with <w> or <x>? (No), does the word taking the suffix end with a single consonant letter? (Yes), is there just one vowel before that? (Yes), is the word taking the suffix a monosyllable? (Yes). The direction is to double the last letter before adding the suffix. But we know that propper is incorrect. 

What am I missing?


Thank you!


Comments (3)

This comment was minimized by the moderator on the site

Oh -- one more thought that I use with this question (that comes up a lot -- and that I often provoke myself) is that there actually is a word spelled "propper" that does have the base <prop> for "to prop something up.

I could have a device to prop up a tablet that is starting to fall apart and I could say, "I need a better propper upper!"

Now my computer does not recognize that spelling - but it is clear that this is the spelling that we should use as the suffixing conventions apply! I suspect no-one had any difficulty reading and understanding the word "propper" in that sentence -- and that it's distinct spelling from "proper" reduces and confusion by following the homophone principle that whenever two words can have the same pronunciation, wherever possible they are spelled differently to mark that difference.



Peter Bowers
This comment was minimized by the moderator on the site

Excellent question Deb and, and equally excellent response Matt!

Deb, one great thing about your question is that you gave so much detail about your thinking that it was easy for the reader to understand your question.

One thing that I recommend as a general practice whenever posing a morphological question is that you ALWAYS include attempted word sums to show your hypothesis.

In this case, your hypothesis was that the word 'proper' for "behaving in a proper manner" had a base spelled <prop> with an <-er> suffix. The left side of that synthetic word sum would look like this:

prop + er →

As you note, when you follow the Big Suffix Checker (find here:, the conventions tell us that if that were the structure of "proper" we would have to double the <p>.

prop(p) + er → *propper

I put that asterisk to show that this is NOT the attested spelling of the word we were trying to construct. I know you gave us all that information in text, but I find it better for my own thinking and that of anyone I share my thinking with if I present my thinking with word sums.

Your question "What am I missing?" at the end of your post is excellent. You identify the details of your hypothesis, how you tested that hypothesis, and identify the fact that your word sum doesn't work and you are stuck.

This is what scientists do -- identify the edge of their understanding, explain that thinking to the scientific community and see if someone can bring more clarity.

I would argue that by doing all of this you have created to opportunity for yourself and others to gain one of the most central lessons of SWI - which I encourage people to think of as 'scientific word investigation' (even though it officially stands for structured word inquiry).

That lesson is to let constructing word sums - which includes application of the suffixing conventions - falsify your hypothesis.

I may be mistaken, but my sense of your statement at the end, "What am I missing?" was suggesting that you still thought that <prop> was the base of "proper", but that you were missing something about the suffixing changes.

It turns out that, in fact, your word sum construction was doing exactly what it is supposed to do. It was falsifying your hypothesis that <prop> was the base of "proper".

The joy of falsification is that it keeps us from banging our head against the wall trying to force our hypothesis to be right despite evidence that it is false. And that frees us up to step back and say -- OK, so the 'structure test' of a word sum falsifies my assumption that <prop> is the base. That means I should stop trying to force my hypothesis and look for another understanding.

If you have a copy of the "4 questions of SWI" (hopefully attached in this response) you will see that Matt took you to the next step. When we are looking to see how words are related, we need to keep in mind that they may be related in two ways.

1) Words might be morphologically related. This means they share a present day English base element. But in order for that to be true, they ALSO must be etymologically related. That is, to share a base, words must share an historical root.

2) Words might be etymologically related but NOT morphologically related. That is to say, words might share the same historical root (and therefore some kind of spelling and meaning connection) but not share the spelling of the same base.

For an example of that second type of relation, note that "please" and "plea" share the historical root placere for "to be acceptable, be liked, be approved" but they do not share a base that can be analyzed with a word sum. The 'meaning test' is 'passed' and it makes sense. I can say "I made many pleas for help" or "please help me" and we can see how close in meaning and spelling these words are.

But the 'structure test' shows these words do not share a base.
You could try this word sum:

*plea + se → please

What is the problem with that word sum? I have no evidence of an <-se> suffix. If I can't make a word sum work, I just have to conclude that the words do not share a base. But in this case the words do share a root. So the words 'plea' and 'please' can be described as 'associated bases'. They are associated with each other through their association with the same historical root.

In this case, Matt pointed you to the key other part of your question -- understanding available through etymology.

Note that he showed the etymology of "proper". If we look at the etymology of the word 'prop' we will see at least 3 free bases spelled <prop>, but none of those are of the same origin as "proper"

So we don't have evidence that "proper" is built on one of the free bases with this spelling.

One could ask, "perhaps there is a bound base <prop> that would not be listed in Etymonline".

That is true! However, your structural analysis with the help of the Big Suffix Checker made it clear that this hypothesis cannot be true because the 'structure test' of the word sum rejected it.

It is so tempting to see that final <-er> as a suffix. Especially when <prop> seems like such a familiar morpheme. But since we are doing science, we need to let the scientific process falsify our favourite hypotheses!

The only conclusion that makes sense to me is what Matt's suggestion was -- that I have no evidence that I can analyze the word "proper" in English, so I treat the whole thing as a base.

Finally, I want to highlight that the key learning from this excellent investigation is not about the word "proper". By far the richest learning this investigation can provoke is the recognition of the fact that we need to be ready to let the word sum falsify our hypotheses even when we are really certain they were right in the first place. It is never a problem to not analyze fully (incomplete analysis), what we need to do is avoid analyzing in such a way to violate the structure of a morpheme and/or the suffixing conventions (false analysis).

With this investigation clear in mind, I recommend going to the page on my website about understanding morphological and etymological relationships and the 'structure and meaning test' at this link:

Thanks so much for sharing your questioning here. I will be pointing others to it as a model of the learning that can result from posing scientific questions!


Peter Bowers
This comment was minimized by the moderator on the site

Ok, I'll take a stab at it, though I hope someone more knowledgeable will jump in soon to correct me. It's a great question, and a bit confusing.

Looking in etymonline, I see "directly from Latin proprius 'one's own, particular to itself,'" which led to the original English spelling < propre > after dropping the Latin suffix < -ius >. I know other English words, such as < theatre > and < metre > where (at least in American spellings) they have reversed the last two letters. The < e > is required in each to make a syllable in English.

So, I'm guessing that < proper > is actually a simple base. If not, as suggested in etymonline, that the < pro- > is a prefix, then the base would be < per >. I'm not sure which one is right, but in either case there would be no consonant doubling.

Comment was last edited about 4 months ago by Matt Matt
There are no comments posted here yet