The previous article looked at “chinjutsu” — the mechanism by which a sentence’s final form determines the character of the whole. This article looks at a phenomenon unique to classical Japanese, in which a single particle in the middle of a sentence reaches across it and determines that final form from a distance.

Add “zo,” and the ending changes

Compare these two classical Japanese sentences.

The second sentence has a particle, “zo,” inserted partway through that the first one lacks. And with it, the sentence’s ending shifts, from “keri” to “keru.” The particle “zo,” sitting well away from the end of the sentence, changes the inflected form at that end — as if operating it by remote control.

The same thing happens with “koso.”

Here the ending isn’t “keri” or “keru,” but yet another form, “kere.”

This phenomenon has long been called “kakari-musubi” — “linking and tying.” Particles like “zo,” “namu,” “ya,” and “ka” link (kakaru) from somewhere in the middle of the sentence toward its end, and the end responds by being tied (musubu) into a specific form — “zo,” “namu,” “ya,” and “ka” tie the ending into the attributive form (rentaikei); “koso” ties it into the realis form (izenkei). This is a classical Japanese rule. It works almost as if a signal, sent from somewhere partway through the sentence, only clicks shut once it reaches the very end.

hana zo saki- -keru (attributive) "Zo" links to the sentence-final "attributive form" in agreement
How "zo" determines the inflected form at a sentence's distant end. Diagram: ToL

What makes this such a strange mechanism

What’s distinctive about this phenomenon is that an inflected form is usually determined by the word immediately before it. But in kakari-musubi, a particle sitting somewhere in the middle of a sentence — sometimes several clauses away — determines the inflected form at the very end. It’s a kind of “long-distance agreement,” a somewhat different animal from the ordinary inflection patterns found in most languages.

And this mechanism has almost entirely vanished from standard modern Japanese. Say “hana koso saita” in modern Japanese, and the ending “saita” doesn’t change at all — the classical-style “tying into the realis form” no longer happens. Why such a striking mechanism disappeared is a large question bound up with the history of Japanese itself, and later articles in this feature return to it.

Summary

PointDetail
The observed phenomenonParticles like “zo,” “namu,” “ya,” “ka,” and “koso” change the inflected form at a sentence’s end
The rule“Zo,” “namu,” “ya,” and “ka” tie the ending to the attributive form; “koso” ties it to the realis form
What’s unusualA “long-distance agreement” — the form is set by a distant particle, not the word right before it
Difference from modern JapaneseThis mechanism has almost entirely disappeared from standard modern Japanese

In a nutshell — classical Japanese had a rule by which a signal sent from somewhere in the middle of a sentence only took shape once it reached the very end. That rule is kakari-musubi.