Pathway to Japanese Literature

The reason I break down the structure of words is that I believe research into words will ultimately lead to grasping the essence of Japanese literary composition.
To consider root inflections, examining compound words first serves as one approach.
For this purpose, one must consider both the root and what is termed the main component of compound words.
When there exists a word such as "mountain," a modifying word combines with it to form a compound word.
As for this main component, I will set it aside without addressing it for now.
I intend to proceed with the discussion by leaving these matters in their tentative state, as doing so poses no hindrance.

Now, as for the main component within compound words, a certain root attaches here and proceeds to form compound words.
In other words, roots attach in a modifying capacity—this being the principle.
Regarding this method of attachment, from our modern perspective, it seems predominantly considered that even in ancient times—just as in the present—the modifiers forming compound words were placed before the main component.
In fact, there were plenty of such examples.
However, when advancing our consideration one step further, we find that in ancient times there existed numerous instances where roots serving a modifying function were positioned below the main component.
On the contrary, that approach would likely have been considered the formal standard to such an extent.
The national language forming the foundation of our written and spoken discourse was recorded after passing through such an era, with traces of these earlier compound word practices remaining into the documentary period.
For example, they referred to ladders as *hashidate*.
When examining the Harima Fudoki, we find that when constructing a ladder to ascend to heaven by stacking straw bags, they wrote *hashidate* for "ladder."
A bridge is a ladder.
Insofar as we understand it today, when one speaks of *hashi*, we think only of bridges suspended horizontally, but in ancient times—whether for crossing or facilitating crossing—anything connecting two places from here to there was called *hashi*, including vertical structures.
They called this *tatehashi*, and when making it into a noun term, they referred to it as *hashidate*.
From our grammatical consciousness, this should be called *tatehashi* (竪橋, vertical bridge), but one must not interpret this as a standing bridge.

The next example shows that while the place name in Yamato is documented as the oldest, terms like Katayama (or Kataoka; alternatively Katayamakayama) were also used as common nouns in Yamashiro and other regions. From contemporary linguistic sensibilities, this would be phrased as "the hill located in the plain beside another hill, which is itself beside another area." There were villages and fields, and accordingly there was a hill there. However, this represents modern understanding. Later, the term came to directly indicate the hill itself, causing people to perceive it as "the hill that stands beside [another hill]," thereby returning once more to simply meaning a hill. Thus, while one might think Katayama ("Beside-Hill") modifies "hill" as a hill's name, people of old—whose grammatical consciousness differed from ours—left components like "land" or "field" implied when forming such terms. This resulted in such structures. In other words, it could equally signify "land beside a hill." Katayama was not a hill's name but meant "beside a hill"—equivalent to what would now feel unstable as kyūbō ("hill-beside"). This follows the same pattern as hashidate: where we would say tatehashi ("vertical bridge"), they used hashidate, and where we would say kyūbō ("hill-beside"), they used Katayama. In words like okabe or okabi, they typically employed the character 辺 (be), though be/bi (or mi) do not mean "vicinity." Therefore, one must avoid translating Katayama with phrases like "hill vicinity," as place name usages differ from this principle. Such words persisted into the documentary era like scattered strata. Simultaneously, remnants of old compound formation consciousness remained—even while knowing that new-era compounds required modifiers above main components—allowing us to discern faint traces of ancient grammatical awareness.

In the language of daily life in the Heian court and among the nobility, there existed the term *shitagutsu* (韈).
This underwent phonetic change and came to be called *shitauzu*.
This likely arose from an awareness of ancient compound word formation.
A blind hung down at the front of the carriage; if raised, the master’s face became visible.
For this reason, there was a lower blind.
This was called *shitausu*.
From our perspective of compound word formation, this phrasing remains somewhat difficult to grasp.
It was, so to speak, “blind-lower.”
When one examines terms such as *lower shoe* (*shitagutsu*) and *lower blind* (*shitausu*), it appears that while *shoe* or *blind* seem like main components with *lower* modifying them—leading one to view them as standard compounds—this cannot actually be asserted.
The truth lay in “lower” occupying its proper position afterward.
At the same time, just as with *Katayama*, it was anticipated that people would universally grasp this implied structure where “lower” followed its referent.
This new-era compound retained vestiges of ancient formation methods—whether through intermittent revival of forgotten patterns or survival through archaic remnants—though no simple explanation suffices.
Such cases might seem novel due to roots appearing first but were fundamentally ancient in method.
This form proved abundant rather than exceptional; I merely cited representative examples rather than rarities.
Even today, meticulous examination of dialects preserving archaic forms would yield countless parallels—particularly among descendant clans diverging before Japan’s documentary era like Okinawa’s people whose language diverged early yet shares core grammar with ours despite stylistic differences—unlike Korea where relations grow more tangled—though undeniably Okinawan remains fundamentally a Japanese dialect despite its antiquity.

However, it diverged at a very early stage.
As our senior colleagues have considered, even if Japan shares grammatical systems with [Okinawan] to the utmost degree, we must acknowledge that differences in style exist.
The reason why there are also many aspects differing from our ancient and medieval grammatical systems is that they diverged before the documentary period.
However, certain types of grammar and word formation methods are completely consistent.
In Japanese literature as well, materials particularly special to the field of Japanese linguistics are not particularly numerous.
Therefore, for the Japanese national language—and particularly for the study of ancient grammar—the Okinawan language is of great importance.
We have no choice but to use this as a supplement and combine it with even the fragmentary remnants of ancient Japanese that remain, working together to attempt to reconstruct a unified form.
Attempting to do so anticipates various effects.
As part of this endeavor, we also seek to examine ancient compound word formation methods.

There were many scholars who criticized Dr. Kanesawa’s theory of Japanese-Korean linguistic common ancestry—a study premised on comparing archaic Japanese with modern Korean—arguing that such a hypothesis was untenable even when viewed against other language family relationships.
However, those who said that were mistaken.
It was common sense arguments—incapable of considering folk traditions’ distinctive traits within linguistic frameworks—that had led to such conclusions.
The fact of linguistic transmission must be considered something possessing power transcending time and space to a certain extent.

The Okinawan language differed from village to village.
This was due to differences among their village deities.
Villages that exchanged words only out of necessity were scattered about; through such isolation they preserved archaic vocabulary.
A significant portion remained governed by divine utterances from these deities.
Traditions such as *omoro*, *otakabe*, and *miseseri*—or related practices—persisted in rural communities.
These continued governing regional dialects well into modern times while maintaining everyday speech in an antiquated form.

The Omoro Sōshi was a work that had collected and accumulated such omoro chants from six hundred years prior in the central regions.

In the Okinawan language, words denoting smallness followed the main component.
This resembled the practice prevalent from Kanto to Tohoku regions of actively suffixing -ko.
In expressions meaning "small something," they used forms like hashigwaa ("small bridge") and ushigwaa ("small cow").
This constituted a structure of [modifier]-following-[main noun].
While resembling Japanese suffixes in form, this did not inherently carry diminutive meaning.
In standard Japanese too, prefixes like ko-, sa-, and wo- indicated smallness; however, -gwaa conveyed affectionate nuance—a term used when approaching something with modest deference, regardless of actual size.
In Okinawa, this compound formation method stood out immediately.
Such grammatical structures proliferated throughout Okinawa, preserved both in documents since the Omoro era and in living dialects.
The archaic form of -gwaa had been -gama.
The term hahagama ("mother-gama"), expressing longing for mothers and ancestral homelands, had transformed into angamaa.
The court ladies' title amushirare meant "acknowledged mother."
The spirit deity Kinmamon represented nothing more than "the true essence of lord-father."
Such phenomena extended to Taiwan as well.
Even limited to Taiwanese languages documented in indigenous tribe surveys—when organizing the speech of "raw" and "cooked" savages—not only vocabulary but grammatical systems too varied between tribes.
Some communities retained elements of Ura-ru aru-tai type structures, though most commonly exhibited so-called South Seas-style grammatical systems.
Certain languages possessed forms resembling English, wholly unlike predicate-final structures.
Linguists had traditionally constructed language families based on word order—an approach amounting to formalistic obsession that would inevitably require correction.
The indigenous tribes differed from Japan and Ryukyu in word order.
Extending consideration from Japan's southern mandate islands to the Dutch East Indies revealed seemingly divergent word orders at surface level; yet superficial analysis suggested numerous foundational consistencies.

However, this was not an immediate issue at hand, and I lacked sufficient knowledge on the matter.
Moreover, when examining the languages of these ethnic groups, they still followed the same form as Japan’s ancient compound word formation methods.
Just as the word order in sentences differed, the method of forming compound words appeared entirely different from that of the modern national language.
But we of old did possess that method.

However, beyond such methods of compound word formation—that is, forms where the modifier preceded [the main component]—there still existed various other ways to create compound words.
One should avoid discussing this in temporal terms; however, since it became markedly apparent in the documentary era, what might be considered a somewhat newer form compared to the earlier one—where the modifying root preceded and the modified main component followed—was, at first glance, an ordinary structure.
That is, because it remained unclear whether two words were merely juxtaposed or compounded together, they adopted a form to distinguish this through a certain method.
If we regard the aforementioned form as old, I believe what came next would be this.
That was first expressed through *akusento*.
Indeed, it appeared that the ancients also made such attempts.
The ancients distinguished between cases where they merely juxtaposed two words—akin to inserting a small punctuation mark (like a comma) in modern usage—and cases where they formed compound words, demonstrating that for compounds, they connected them below using *akusento*.
While the details remained unclear to me, only in the Kojiki did that exist, albeit sparingly.
In traditional proper nouns that should be revered or in mystical phrases, this method had been adopted.
It remained unclear whether compound words prompted *akusento* or whether *akusento* enabled compound words to fulfill their function; regardless, this consideration could not be dismissed.

In connection with this matter, it became necessary to pay attention to root inflections during compound word formation.
Traditionally speaking regarding these nominals and nouns' inflections—little had been discussed.
Yet this matter held crucial significance.
While nominal endings remain fixed today evidence suggested they had shifted fluidly in antiquity.
Such instances abounded abundantly.
He commenced by examining phenomena conventionally termed ordinary phonetic shifts.
When nouns bearing e-column vowels formed compounds they transformed into a-column counterparts.
Consider *sakedaru* becoming *sakadaru*, *kazeboroshi* shifting to *kazaboroshi*, or *sugegasa* altering into *sugagasa*.
This constituted no mere sound alteration—rather it embodied conventions systematized through natural mechanical organization of root inflections during compounding.
Delving into origins revealed that during compound formation roots initially assumed u-column configurations.

*Kamikaze* was without exception pronounced as *kamukaze*.
The character *sai* had both 'i' and 'yu' sounds, and its phonetic value had been considered to fluctuate; however, this 'i' was an unchanging sound that shifted to 'yu' when forming compound words.
To explain why this phenomenon occurred: when considering the state immediately preceding this transformation, it was found that all roots came to take on what might be called u-column sounds for their final elements—that is, due to their proximity to consonants—such that when they underwent fluctuation, they shifted into the u-column.
Therefore, even when verbs were formed, their terminal forms ended up as u-column sounds.
Among verbs, the most unchanging element was this terminal form.
When roots connected and combined, the u-column sound emerged.

Because roots were close to the u-column, this consideration came first, and their nature came into play when forming compound words. An awareness emerged of proximity to the u-column; unable to be fully satisfied with roots alone, inflections arose. The modifier changed into the u-column. For example, there was the word *kogane* (gold). The term "yellow" was *ki*—though it would not be the yellow we envision—this *ki-gane* became *ko-gane*. This *kogane* must also have been fluctuating. In ancient times, it was called *kugane* or *kugani*. In ancient times, while it was a noun called *ki*, when forming compound words, once people became conscious of the form where the root and main component combined relative to the compound’s main component, the u-column sound would emerge. In our country as well, the i-column and u-column were close. People had been saying "ku" for "tree" all along. The deity of trees was called Kukunochi no Kami. The deity of gourds was called Kuhizamochi no Kami. Kuhiza referred to a gourd made from wood. There was one reason for *ki* changing to *ku* as well. It was that the roots reverted to the habit of forming compound words. When forming compound words, fire was called *ho* almost without exception. *Ho* and *fu* were nearly a single sound. Uto and o were close, but ho and fu were even closer. This could not be dismissed as mere phonetic change. With such facts existing and unconscious awareness arising—leading to the application of these rules—it came to be subsumed under what was termed "phonetic change."

This constituted the second stage of compound word formation, becoming what was termed the inflection of nominal endings.
In other words, predicatives underwent inflection independently, whereas the inflection of nominals involved a main component that continued below.
This was an important matter when considering nominals and predicatives.

Next came a phenomenon we considered to be phonetic change, where what stood out most conspicuously was the inflection of modifier roots into a-column sounds when forming compound words.
That is, cases where the main component of modification took on an a-column inflected rhyme.
Examples included *sugagasa* and others.
Here, there were many cases where what we considered to be independent nouns had omitted the main component of the compound word.
It could be roughly surmised that “white hair” had undergone inflection as *kegaka*.
However, when the hair’s direction was above and modified the main component, it was clear—but why did it become so when below?
Since this meant “white-haired hair,” there had to be another main component of the compound word below this.
The parent/ancestor term undoubtedly derived from *oyu* (“old”), and I believed it originated either from the verb *oyu* or from a pre-verbal root—something like *oyu*—that predated verbs.
One could anticipate a term such as “old person” below.
Even without explicitly stating “person,” it implied this.
This became possible to comprehend even when disassembling such terms because they had undergone inflection.
*Oyu* merely signified being aged—that is, being elderly.
Since the elder managed the household, in ancient times, the *oya* was primarily female, and the ancient *oya* in the matriarchal era had been *haha*.
Later, it came to refer to males.
In ancient times, *haha* was written with the character for “ancestor.”
The character 祖 could denote ancestors in some cases, but it often carried the meaning of “mother.”
Mioya-gami also held significance there.

The rope undoubtedly originated from the word *nau*. Initially, ropes used wild plants in their natural state, requiring almost no twisting. They called them *tsuta*, *tsura*, *tsuru*, and such. To make them thick and strong, they had to twist strands together. Thus twisted vines came into being. Alternatively, they used *shimoto* shoots by twisting them. *Shimoto* refers to the straight new growth from shrub branches. They would say phrases like “not meant to be twisted into thread.” They distinguished by size—small ones they called *yoru* (twisting), large ones *nau* (plaiting). While the exact meaning remains unclear, “rope” signifies something plaited. *Nau* is thought to have shed sounds from forms like “object” or “plaiting vine,” coming to be expressed simply as *nawa*. From this too we see that changing sounds to the a-column for compound formation stemmed not from mere phonetic change.

Next come i-column compound words.
These examples are extremely numerous and have come to be regarded as commonplace facts.
Both their modifying elements and main components retain their own distinct vitality.
When nouns combine with verbs or verbs with nouns—whether in noun-verb or verb-noun pairings—both *tori* ("taking") and *sashi* ("inserting") persist as *tori* and *sashi*.
Though two words bind together to form one compound, they continue to exist separately.
This applies equally to terms like thought-matters and gait-steps.
Yet there remains no doubt these constitute true compound words.
This represents an exceedingly straightforward method of compound formation.

In compound word formation methods other than those shifting into the i-column, compound words in antiquity were more frequently formed from the attributive form rather than the ordinary i-column continuative form. Compound words derived from the attributive form might give one the sense that their status as true compounds felt incomplete. For instance, terms like *moyuru hi* ("burning fire") or *izuru yu* ("welling hot water") would prove difficult to recognize as compound words. While structurally compounds, they failed to fully convey the essence of nominals. Yet in reality, this form had been prevalent in earlier times. The literary period represented an era when compounds formed from continuative and attributive forms contended for dominance, with evidence suggesting that the u-column attributive method derived from i-column roots was actually older. This method, as previously discussed, developed into a pattern continuing from the u-column to the main component. However, the perceptual awareness of this u-column continuation gradually shifted over time. This transformation became instrumental in forming both terminal and attributive forms of predicatives, with the fourth conjugation emerging directly from compound formation processes. This point proves critical for understanding how Japanese predicative conjugations originated—while *izuru yu* ("welling hot water"), an attributive-form compound, lacks definitive compound-like linguistic tonality, it predates *ide yu*. Ancient speakers inherently perceived this distinction. To rephrase: i-column continuative compounds maintain a sense of semantic separation between main component and modifier while achieving perfect alignment. By contrast, attributive-form compounds not only retain this separateness but also imbue their main components with distinctly verbal qualities.

To discuss root inflection, it was necessary to address compound words first.
Therefore, I concluded that focusing on the compound formation methods of the a-, i-, and u-columns would suffice for this initial analysis.

Next，I would like to attempt to explain，to the extent possible，how verb conjugations came into being．In linguistic studies，when one pursues considerations beyond a certain point，they inevitably become speculative；therefore，one must stop at an appropriate juncture．I have previously explained that roots undergo inflection before compound words form．Moreover，you should have understood by now that nominals—which share their form with roots，being subsumed under noun forms—likewise undergo inflection．

Now，when we attempt to connect this with the origins of verbs used today，what becomes clear is that verb genesis remains fundamentally undeciphered．This stems from our language’s traceable past containing only forms that achieved a certain degree of development．Nevertheless，I maintain that envisioning an era when roots operated more freely may cast faint light on how verbs and predicatives emerged．

There are two pathways here．The first involves direct inflection of individual nominals；the second entails forming compound words to attain relatively complete predicative forms—that is，these two approaches consist of either immediate nominal inflection or passing through compound word formation to create predicative structures．

The term *furu* blurs boundaries between nominal and predicative．This predicative-like *furu* simultaneously exhibits nominal qualities．Even without forming compounds，it clearly performs nominal functions．This generates inflection．Within known forms—*fura*，*furi*，*furu*—we observe structures closely resembling quadrigrade conjugation．Meanwhile，forms like *fururu* and *furure* remain incomplete．Those verbs derived directly from nominals functioned only through forms convenient in particular historical periods，with others falling into disuse．Many such cases exist．Even regarding *furu*，I consider forms predating its attributive form dubious．

The word *fuyu*, that is *fueru*, did not merely signify proliferation but also division—the splitting into entities sharing the same nature. While *fuyu* now functions solely as a lower bigrade verb in our understanding, its connotation of division had been more prevalent than proliferation in antiquity.

“Prince Ōsazaki of Honda’s Day, the great sword he wears.
Base sword, tip splits...”

When the *Ōjin-ki* recorded this, it described a double-edged sword whose tip branched into multiple divisions—a blade with bifurcations where the end split apart. Since *fuyu* involved soul-division, it served as a type of *kotohogi* ritual incantation. When this *fuyu* clearly lexicalized as a noun, it became the season of winter—this transformation remains beyond doubt.

At each year’s end, communities performed the *mitama no fuyu* festival to split souls. That ceremonial period itself was called *fuyu*. From this ritual usage emerged an extremely contracted form that eventually solidified as the seasonal term “winter.” By analyzing the transitional phase of the *mitama no fuyu* festival, we discern how the *fuyu* in “*fuyu* festival” gradually acquired nominal qualities.

The examples cited by grammarians mixed ancient and modern elements.
As a result, there were instances where it remained unclear whether these represented actual examples or metaphorical usages.
Therefore, I avoided examples from the early modern period in my discussion.
For instance, requiem songs were termed *tamafuri* songs.
In various regions, these requiem songs manifested as *kuni-buri*.
Over time, they gradually lost their original significance and assumed the character of what we now call folk songs.
For *kuni-buri* to acquire the meaning of "the nation's *tamafuri* songs," considerable time must have elapsed while the concept matured in people's consciousness.
The term *fuyu* follows this same developmental principle.
Merely uttering *fuyu* fails to conjure our modern conception of winter.
Only through understanding the profound historical reality behind this word did its simplified form *fuyu* emerge and become fully comprehensible.
When assuming compound word structures, their forms remain relatively distinct.
The ritual term *mitama no fuyu*, denoting soul division through ceremonial practice, maintains comparatively stable linguistic form.
Later this evolved to signify receiving divine beneficence.
Even as roots gradually assumed predicative characteristics, they operated with latent expectations of forming compounds.
True terminological completeness proves unattainable unless we recognize compound words existing independently from their modified principal components.
This stems from their deficiency in carrying implicit meaning as autonomous lexical units.
While root inflection and compound adhesion create terminology, inflection inherently presupposes compound formation.
The strongest manifestation of this principle appears not in inflected verbs/auxiliaries/adjectives, but in predicatives bearing suffixes.
These suffixes differ fundamentally from grammarians' conjugational markers.
Through etymological analysis, we discern divisions along semantic boundaries.

While some scholars insert a stem between root and suffix, I set aside such arguments here to focus solely on roots.
When considering suffixes in their terminal forms - despite minor variations - all employ u-row sounds.
The theory that *ku*-ending words mean 'come', *nu* signifies 'go', and *ru* denotes 'exist combined' holds factual recognition within limits.
Examining works like Man'yōshū where Japanese uses Chinese characters reveals *ku* and *keri* written with 来 ('come').
This reflected period-specific understanding through sinographic representation - reasonably coherent methodologically.
Moreover, all contemporaries shared this interpretive approach.
Yet certain meanings resist decipherment.
While monosyllabic *ku* plausibly means 'come', its function in *kaku/toku* remains opaque.
This phenomenon stems from root inflection but had been explained through anticipatory unification - the so-called 'one sound-one meaning' phonosemantic theory.
Pushing further led analysts seeking meaning in vowels/consonants themselves.
Tachibana Moribe demonstrated lexical expertise but remained mired in this mindset regarding suffixes.

Just as prefixes are meaningless, so too are suffixes meaningless, existing merely to manipulate roots—this line of thinking must be discarded, I believe.
To overly attribute meaning or to deem something meaningless—neither is advisable.
The awareness of roots such as *i*, *nu*, or *ku* cannot be said to be entirely mistaken, for this awareness emerged after a certain period as an organized system of such suffixes; in other words, it is best viewed as the form suffixes took after a specific era.
The relationship between roots and suffixes must also be deduced from the form of roots when forming compound words.

I will now take as an example words that carry a relatively modern feel.
There exists a theory that *minoru* derives from *mi ga noru*.
For us, the sense that this word constitutes a phrase or sentence has atrophied, while its verb-like quality runs deep.
Though perceiving *tagayasu* as a single word, I cannot suppress the sense that it means "turning over fields."
Therefore, even when considering verbs emerging from compound words, one discerns a trajectory whereby the feeling of two conjoined words gradually diminishes.
However, when expounding upon the origins of ancient predicatives, it proves ill-advised to presume they developed from forms like *minoru*—that is, from structures such as *mi ga noru* that retain a sentence-like quality.
I contend we must perceive between root and main component not an intimate relationship but rather a psychological leap.

In many predicatives, particularly verbs, the main component remains small while the root grows large.
Moreover, this main component determines the very function of verbs themselves.
That is to say, it is through conjugated forms that verb structures become defined.
In my view, verbs constructed from roots and main components do not represent primordial forms; their immediate predecessors arose through inflections emerging directly from roots.
While *iku* is understood today as meaning 'to live' or 'to breathe,' its root usage appears in terms like *iku bow* and *iku arrow,* conveying a sense of latent potency.
When assuming adjectival form, it manifests in constructions like *ikashi.*
This state of root inflection inevitably governs even cases where verbs develop from secondary compound formations.
These are no mere simple compounds.
What we call verbs must undergo transformation from simple to complex structures—forms like *minoru* cannot emerge without substantial evolutionary progression.

The meaning of u-row suffixes arose from more semantically complete words being broken down—that is, from the practice of inflecting nominals into verbs—and I believe they represent older meaningful words that were dismantled and fossilized.
If I were to put it that way, there would be a slight risk of misstatement.
The terminal forms *uku*, *sutu*, and *nu* belong to the u-row, but the terminal form itself emerged last.
When viewed through adjectives, this becomes clearly understood.
It simply cannot be considered that they began from the terminal form.
The continuative and attributive forms first came into being, and from there developed a tendency for the terminal form to emerge.
When it comes to secondary elements, one cannot consider them as having begun from what appears deceptively simple—the terminal form.
The origin of verb endings cannot be discerned through isolated study of u-row suffixes; one must examine them comprehensively across the entire conjugation system.

Therefore, using old mistaken theories, I intended to re-examine the matter once again.
It was a theory that could be accepted in part but had to be discarded in its entirety—one that posited verbs inflected via the form of nouns.
To give an example, there were several elements within the word *nagameru* that shared the same pronunciation but had different origins.
That is to say, homophonous words with distinct meanings existed.
Among these was *nagame*, a term exclusively employed during the Heian period to denote the melancholic state of lovers kept apart.
*Nagame* encompassed both gazing at distant vistas and chanting with sigh-laden breaths.
Such cognate terms appeared to gradually accumulate layered significances.
This *nagame* fell precisely into the theory I had previously rejected.
It was used when describing sexual melancholy or lovesick yearning.
The *Kojiki* records: “He was made to undergo perpetual *nagame*, remaining unwed and sunk in brooding—(old reading).”
*Me* signified the meeting of man and woman; when its nominal sense intensified through usage, it came to mean “wife.”
The phrase “nagame o heshime” indicated married couples refraining from intimate discourse due to prolonged separation.
This *nagame* constituted the root of *nagamu*.
Though undeniably valid, I had persistently denied this connection.
While maintaining that nominal forms functioning verbally represented error incarnate, I now recognized the necessity of its consideration.
The concept persisted through vestiges of *fu* (経) or *he* attached to *nagame*.
Rather than direct conjugation from *nagame*, *nagamu* emerged from nebulous awareness of forms like *nagame o he* or *fu*.
That is—though no formal fusion occurred—there existed an intrinsic psychological continuum.
This mirrored winter’s emergence as distinct from the Mitama’s winter festival rites.
Thus did I resolve to investigate verb endings’ developmental trajectory.

Precisely appropriate points could still be explained regarding nagame.
This too was undeniably factual.
Nagame was a ritual abstinence during the rainy season.
During this period, people believed deities manifested in this world.
Satsuki (May) and Nagatsuki (September) were seasons of prolonged rains; alongside September, May had been a time of strict ritual abstinence.
Among these, September had detached from customs to become purely faith-oriented, while May endured through its connection to rice-planting rituals—hence the reason became evident.
In May, people could not meet until rice planting concluded.
This was termed nagame abstinence.
Though married couples, they were forbidden to meet.
This aligned with the phrase “made to undergo nagame... [and] granted.”
Ultimately, which came first remained unclear, but if provisionally settled, nagame emerged from abstinence practices during long rains, from which me developed into phrases like “nagame o heshime... tamau”—yet given our deficient preparedness and sensitivity toward ancient language, we could not conclusively determine this.

When a single word was created and put into use，an understanding of other similar words became activated；that is，a secondary etymology came to overlay the original word．What caused this was the so-called folk etymology theory．When that happened，it became unclear where［it］had originated from．And thus，terminological examples continued spreading infinitely．The verb *nagamu* undoubtedly originated first from the noun *nagame*．Whether “long rain” or “nagame，” they are nouns．That *nagame* did not emerge solely from root inflections but had come about through a process of verb-like psychological transformation．That was to say—if I may put it that way—the sense of what might be called “performing *nagame*” (*nagame o nashiageru*) exerted a powerful influence，yet it did not participate in taking the “*-mu*” ending．Nagame and nagamu were clearly distinct，but when we brought in etymological theory，there existed a single thread of connection between them．When it came to secondary meanings，there were also cases where roots and suffixes clearly separated．In any case，it could not be considered that verb endings first created u-row sounds．Thus，first the continuative or provisional forms—and rarely the attributive form—came into being．

Here, I must address one more matter: the issue of the driving force that gave rise to the terminal form of verbs. It is the mind that unconsciously senses a certain tendency and seeks to position itself under that tendency. The language we use was not definitively standardized into categories such as lower bigrade, upper bigrade, quadrigrade, and irregular conjugation; however, through gradual organization, several conjugation forms came into being. The endings of each verb were pruned and gradually unified into a certain form. While generally fixed into eight conjugation types, if one considers phonetic changes, they became much simpler. Even verb conjugations were pruned and fixed into unified forms. One trend to regard is that the terminal form and attributive form converged into one—something formed through sound pruning. When examining this example, the endings take on both simple and complex forms. In quadrigrade verbs, this distinction remained unclear, but when observed in lower bigrade verbs, the terminal and attributive forms separated and became complex. That which was pruned and refined into a simple form became what we call quadrigrade conjugation. However, words—even in such contexts—involved circumstances where conjugation arose from other points. Thus conjugation forms gradually became organized; however, older terminal forms often took a form resembling the attributive form. If we ask what pruned them, it would be the endings formed through reduction of main components within compound verbs—between main components and roots—and we may infer these endings certainly held meaning in ancient times. Though it was more complex than what people today conceive, be that as it may, the terminal form may be said to indicate the path along which suffixes gradually developed.

(I stand upon a different vantage point, unable to find full satisfaction in employing the so-called terminology related to verb conjugation stages and forms as they are conventionally used.)
However, in this brief treatise, I resolved to eschew the approach of tediously enumerating fastidious new technical terms obsessed with rigid nomenclature while feeling compelled to provide exhaustive explanations for each.

At the same time, I must apologize that this work derives from hastily compiled notes; upon rereading it, I was startled by how it has taken on a didactic tone and feel I have been discourteous to the esteemed readers who examine such writings.)