Standardization of Finnish orthography: From reformists to national awakeners

Full text

Finnish
Standardization of Finnish orthography:
From reformists to national awakeners
Taru Nordlund
0. Introduction
This article examines the standardization of Finnish orthography, concen-
trating on the two periods of rapid development: the 16th and the 19th cen-
turies. The first part discusses the role of the Western Church at the early
stages of written Finnish: conversion to Christianity and especially the Lu-
theran Reformation in the 16th century set the foundations for the usage of
Finnish as a literary language. To provide a background to the standardiza-
tion of orthography in the 16th and the 19th centuries, the spelling system
of Modern Finnish is introduced in section 2. The following sections 3–5
discuss the orthography of the earliest writings and provide some examples
of the problems that the first writers were faced with. The second external
trigger for the evolution of Standard Finnish was European nationalism,
which had reached Finland by the beginning of the 19th century. This era
was characterized by a strong desire for nation-building, which was linguis-
tically reflected in the process of standardization with its often heated de-
bates about the “authenticity” or “purity” of the language. In section 6,
aspects of 19th-century spelling are discussed. In most respects, the orthog-
raphy of 19th-century Finnish resembles that of Modern Finnish. However,
there were some debates on orthography that clearly reveal the nationalistic
aspirations of the time. At the end of the article, one of these debates is
discussed in more detail.
DOI 10.1515/9783110288179.351,©2017 Taru Nordlund, published by De Gruyter.
This work is licensed under the Creative CommonsAttribution-NonCommercial-NoDerivs3.0 License.

352 Taru Nordlund
1. The role of the Western Church in the early stages
of literary Finnish
1.1. Traces of written Finnish in medieval Finland
In the 12th century, the Catholic Church stretched its northernmost point of
influence to Finland. However, the language used by the Church, Latin, had
no tradition in Scandinavia. It is therefore difficult to estimate how much of
the religious message of the Church was meant to be taught to the people,
and more importantly, how much of this message was actually understood.
Even though the language of the Church was Latin, it has been speculated
that some Finnish must have been used in the Mass and in everyday parish
life, even in Catholic times (Maliniemi 1955: 82–112).
Before the Reformation, no documents exist in written Finnish. How-
ever, the statutes of the Synods shed some light on the linguistic situation in
medieval Finland. These statutes explicitly set down the most important
doctrines and teachings that were to be explained for the people in their
own language. The Synods of 1441 and 1492 are especially important in
this respect. These Synods, including as their delegates the bishops of
Turku, Maunu Tavast and Maunu Särkilahti, ordered that the basis of popu-
lar education should be established in the vernacular: particularly, the Pater
Noster, the Credo, the Ave Maria and the Modus Confitendi should be
translated into the vernacular (in linguam maternam), read aloud at church
every Sunday, and written down to ensure that their wording remained
stable enough to be learned by the people.1 It is also clear that the sacra-
ments that involved some kind of interaction with the parishioners, for ex-
ample communion, baptism and marriage, must have been partly in Fin-
nish. Thus, even though no written documents in Finnish exist from the late
Middle Ages, it is probable that these early ceremonies moulded the lan-
guage for religious uses and later formed the basis for the literary use of
Finnish (Maliniemi 1955: 82–112).
1.2. The Reformation
As stated above, the earliest printed texts in Finnish date from the first half
of the 16th century. The history of the standard written language has its
origins in translations of the Bible and other religious texts into vernacular
Finnish, the language of the common people. The reformers set to work,
with Latin, German and Swedish orthographies as their models.

Finnish 353
The first authority on Finnish orthography (often described as “the fa-
ther of written Finnish”) was Mikael Agricola, bishop of Turku and the
best-known figure of the Finnish Reformation. In 1536, Agricola was sent
to Wittenberg to study the doctrine of the Reformation under Luther and
Melanchthon. No doubt, Agricola and other young students from Finland
were sent to Germany with the intent of translating the Bible into Finnish.
And indeed, Agricola lived up to these expectations and became a pioneer
in the cultivation of Finnish in written form, by translating no less than
2500 pages, including an ABC-book (an early catechism), the New Testa-
ment and parts of the Old Testament, a book of sermons and several other
books to be used by ministers. Agricola worked methodically, and he usu-
ally used several sources in Swedish, German, Latin and Greek to create a
Finnish text that was adapted to the Finnish context. Agricola’s texts can be
understood by a modern speaker of Finnish, with a small amount of train-
ing in his spelling. Apart from those of Agricola, Finnish texts from the
first part of the 16th century only exist as a few isolated manuscripts, and
after Agricola, almost a hundred years’ silence in written Finnish followed.
This was probably due to the impoverishment of the Church, caused by
royal policies. The first edition of the whole Bible appeared in Finnish in
1642 (Heininen 2007).
Sections 3 and 4 discuss features of Agricola’s orthography. To provide
some background information for this, basic features of Modern Finnish
orthography are introduced in the next section.
2. The orthography of Modern Finnish
Modern Finnish has almost a one-to-one correspondence between pho-
nemes and graphemes: each grapheme corresponds to one and the same
phoneme, and each phoneme corresponds to one and the same grapheme.
As a result, Finnish orthography uses almost as many graphemes as there
are phonemes in the language.2
Table 1. Phonemes and graphemes of Modern Finnish
p
honemes
c
orresponding graphemes
/a/, /e/, /i/, /o/, /u/, /y/, /æ/, /ø/
3
<a>,<e>,<i>,<o>,<u>, <y>, <ä>, <ö>
/d/, /h/, /j/, /k/, /l/, /m/,
<d>, <h>, <j>, <k>, <l>, <m>,
/n/, /ŋ/, /p/, /r/, /s/, /t/, /v/
<n>, <p>, <r>, <s>, <t>, <v>

354 Taru Nordlund
Phoneme length is an essential distinctive feature in Finnish: differences
in length of the sounds very frequently serve to distinguish separate words.
In writing, a long sound is always written with two letters, a short sound
with a single letter (Karlsson 1983: 16–19, Laaksonen and Lieko 2003: 15–
9, Sulkala and Karjalainen 1992: 365–374).
There are very few exceptions to the phoneme-grapheme correspon-
dence. There is only one phoneme, the phoneme /ŋ/, which does not have a
grapheme of its own.4 And indeed, the phoneme /ŋ/ is not a full phoneme,
insofar as it only ever appears as a long sound, and always as a result of
morpho-phonological alternation in inflectional forms, e.g.,
(1) lanka langa-t
[laŋka]5 [laŋŋa-t]
‘thread, wire or string’ ‘thread, wire or string’ + PL
Another feature of spoken Finnish that is not represented in orthography is
so-called consonantal reduplication. After certain morphological catego-
ries, the initial consonant of the following word is lengthened: that is, in the
written form, a word ends in a vowel, even though a word-final consonant
is pronounced. If followed by a vowel, a glottal stop is pronounced:
Table 2. Comparison of written and pronounced forms
writing pronunciation
tule_tänne
[tulet tænne]
‘come-IMP. 2.S. here’
lapsille_leluja
children-ALLAT. toys-PART.
[lapsillel leluja]
‘toys for the children’
mennä_ elokuviin
go-INF. movies-PL.+ILLAT.
[mennæʔ ʔelokuviin]6
‘to go to the movies’
Consonantal reduplication has its explanation in phonological history. At
an earlier stage, the consonants /k/ and /h/ were used in word-final position
in several morphological groups (e.g., *tulek tännek ‘come here’). In the
spoken language, the final /k/ or /h/ was assimilated with the first sound of
the following word. The final /k/ and /h/ were lost in most dialects, but as a
relic from older days, the consonant is even today realized either as a gemi-
nation of the following consonant, or as a glottal stop (see examples above;

Finnish 355
Hakulinen 2000 [1978]: 51–52). Consonantal reduplication has no counter-
part in the writing system. At the end of this article, 19th-century attempts
to standardize the notation of this phenomenon in Finnish orthography will
be examined.
3. The orthography of the earliest writings in Finnish
3.1. Foreign models
As there were no manuscripts written in Finnish before the 16th century,
the closest models for writing were Swedish, Latin and German texts. Until
1809, Finland was part of Sweden, and in addition to Latin, Swedish was
used in writing. Typologically, Finnish differs from most European lan-
guages, and thus the earliest writers faced some serious problems in the
standardization of its orthography.
Table 3. Agricola’s discrimination of vowels in different positions
initial syllable
final syllables
Modern Finnish: all syllables
caatua
loppun
kaatua, loppuun
‘to fall’
‘till the end’
Firstly, there were phonological distinctions that should have been made in
orthography, but were not, as there were no suitable models for them in
other languages. An example of this is the marking of long vowels. For
lack of a model (at least, a systematic one) in other languages, the marking
of long vowels was at first very sporadic. In Agricola’s works there was,
however, a difference between the initial syllable and the final syllables of
the word. Originally, Finnish used to have long vowels in the first syllable
only, and thus all the long vowels in the final syllables are secondary. They
arose as a result of contraction (e.g., *talohon > taloon ‘into the house’). In
the initial syllable, Agricola often used two letters to mark long vowels,
even though marking with only one letter was also possible. In the final
syllables, Agricola usually used one letter only (Häkkinen 1994: 174, Lehi-
koinen and Kiuru 2001: 62–63, Rapola 1965: 54–55).
As we shall see in section 4.1., Agricola’s tendency to use just one letter
in final syllables may have been influenced by the phonological system of

356 Taru Nordlund
the south-western dialects of Finnish where all long vowels in final sylla-
bles are shortened.
The second feature of orthography probably influenced by the source
languages is the tendency to create hypercorrect distinctions, that is, the
earliest orthography sometimes represented phonological features of the
model languages, features that Finnish did not necessarily share. An exam-
ple of this is the variation in the marking of the phoneme /k/, which in Fin-
nish is a palatal or velar stop. In early written texts, the notation of /k/ often
followed the models of Latin and Swedish. When /k/ preceded a front
vowel, the grapheme <k> was used, and the grapheme <c> often (but not
always) preceded a back vowel. The orthography thus reflected the pronun-
ciation of Medieval Latin (and Modern Italian). When preceding the vowel
/u/, the grapheme <q> was sometimes used for /k/, which was also a Latin
tradition. And when /k/ preceded the vowels /e/, /æ/ or /ø/, the grapheme
<ki> could sometimes be used, which in turn reflected Swedish pronuncia-
tion (cf. Modern Swedish känna [çenna] ‘to know, to feel’) (Lehikoinen
and Kiuru 2001: 65). However, it is important to remember that Agricola’s
writings (and those of his contemporaries) always show a great amount of
(so far unexplained) variation in all these cases.
Table 4. Unexplained variation in Agricola’s works
Agricola Modern Finnish
lukea
read-INF.
‘to read’
lukea
caunis
‘beautiful’
kaunis
kieula
‘bow (of a ship)’
keula
quin
(conj.) ‘as, like, that’
kuin
3.2. The phoneme /d/ – an orthographic loan
The phoneme /d/ is not part of the phonemic system of any Finnish dialect.
It was introduced into Standard Finnish in the 19th century as an ortho-
graphic loan from Swedish. In Agricola’s orthography, the graphemes <d>
or <dh> were used for the voiced dental spirant [ð]. The model for the or-

Finnish 357
thography came from Swedish, which used a similar orthographical con-
vention for a similar type of spirant sound (see Zheltukhin, this volume). In
Swedish, /ð/ developed into /d/ in the 17th century, and the orthography
changed accordingly. Consequently, the orthographic variant <dh> disap-
peared from Finnish texts, and, in the 1642 Bible, only the grapheme <d>
was used to note [ð]. The spirant sound then started to disappear from spo-
ken Finnish as well, and as the spirant gradually fell out of use, the pronun-
ciation of <d> as [d] (as an apico-alveolar voiced stop) spread. Thus, a new
phoneme was added to the phoneme system of Finnish, mainly after the
written model (Lehikoinen and Kiuru 2001: 74–75).
Until 1809 Finland belonged to the Kingdom of Sweden, and all au-
thorities, officials, priests, and teachers either came from Sweden or were
educated in Swedish. It is therefore understandable that the pronunciation
of /d/ did not cause any trouble for the Swedish-speaking intelligentsia
(who had the same sound in their own language), but it proved to be ex-
tremely difficult for the Finnish-speaking people. Thus, it is no surprise that
this “newcomer” was the source of a heated debate in the nationalistic lan-
guage reform in the 19th century, as will be shown in more detail in sec-
tion 6.
4. Some traditional and less-traditional explanations for the variation
in the earliest texts
4.1. Diatopic variation: Features of south-western dialects
in 16th-century writings
Since the city of Turku (Åbo) on the southwest coast of Finland was the
capital city until 1812, Standard Finnish developed primarily out of south-
western dialects. South-western features are particularly clearly seen in the
texts of the earliest writers. As mentioned above, even the marking of long
phonemes in final syllables with single vowels in Agricola’s writings could
have been a feature from the south-western dialects. In these dialects, all
long vowels in final syllables are shortened (cf., e.g., Rapola 1965: 70):
(2) ehtoo
‘evening (a dialectal word)’
>
ehto
[monta] kertaa
[many-PART] time-
PART.
‘many times’
>
[mont] kertta7

358 Taru Nordlund
As stated above, apart from Agricola, only a few short and isolated manu-
scripts exist from the first half of the 16th century. One of these is The Gos-
pel Book of Upsala, a manuscript that consists of 31 pages of texts from the
Gospels and the Epistles. This manuscript has a lot of variation in the mark-
ing of long vowels, but it does not display the discrepancy between the
marking of the initial and the final syllables as seen in Agricola’s texts
(Penttilä 1932: 41–43).
Another feature in Agricola’s orthography that probably reveals diatopic
variation is the marking of geminate nasals and liquid consonants. As stated
in section 2, double consonants have a phonemic status in Finnish, and in
Modern Finnish spelling they are always written with two letters. In the
earliest writings, the marking of geminates shows a lot of variation.
However, even though the spelling of geminates was unstable, Agri-
cola’s tendency to write, in particular, double nasals (mm and nn) and liq-
uids (ll and rr) with only one letter seems to suggest that this is not pure
coincidence. This tendency can be observed in the Gospel Book of Upsala
as well. Indeed, the spelling of geminate nasals and liquids is especially
complicated because of a regional south-western feature: in these dialects,
these geminate nasals and liquids were shortened after a long vowel or a
diphthong, or when preceded by a syllable that did not have main stress:
(3) aallot
wave–PL
‘waves’
>
aalot [geminate l shortened when preceded
by a long vowel]
annamme
give–IND.1.PL
‘we give’
>
anname [geminate m shortened when
preceded by a syllable with no main stress]
In Agricola’s texts, the spelling is systematic enough to suggest that the
words written with one letter would have been pronounced as short, and
therefore the variation seen in the spelling of geminate nasals and liquids
can partly be explained as diatopic. In addition, some cases seem to be lexi-
cally governed, so that there are morphological categories or words where
spelling with one letter is much more usual than in others (Lehikoinen and
Kiuru 2001: 82–83, Penttilä 1932: 80–81).
Of course, we cannot know for certain exactly what served as a model
for Agricola in his marking of long vowels in final syllables or of geminate
nasals and liquids. In all probability, the spelling was guided both by for-
eign models and the influence of south-western dialects. And of course, we
do not know the intentions of the early writers themselves: was it their in-

Finnish 359
tention to reflect spoken language, to indicate that the vowels in the final
syllables were really pronounced as short, and geminate consonants as sin-
gle consonants? Or did they want to follow models from other languages?
If so, then why did they sometimes use long vowels instead of short ones?
There was no written standard for 16th-century Finnish, a standard that
would have led to the emergence of a non-variant form. As to some vari-
ables, we will probably just have to come to terms with the idea that there
is no ultimate solution in understanding the variation, and no one single
way of interpreting the texts.
4.2. Are there other explanations for variation?
Apart from diatopic variation, Agricola’s spelling shows diachronic varia-
tion. One example of this is the spelling of the spirant sound [γ] that was
used as a weak grade variant of [k]. It has been stated that the spirant /γ/
was disappearing from Finnish in the 16th century, and the varying pronun-
ciation is probably revealed by the variation in spelling (see e.g., Häkkinen
2007: 70):
Table 5. Agricola’s spelling variants for [γ]
<gh> <g> Ø
roghan
food-GEN
‘food’s/of the food’
algusta
beginning-ELAT
‘from the beginning’
luØen
read-IND.1P
‘I read’
Several studies have pointed out the variation to be found in Agricola’s
different works, for example, the translations of the New Testament from
earlier and later stages of his career. Surprisingly enough, no systematic
research on Agricola’s works has yet been carried out. A morpho-
syntactically coded database of Agricola’s works is being prepared at the
moment, and the first critical edition of Agricola’s works, an edition of the
ABC-book, his first printed text, was published in 2007 (see Häkkinen
2007). Along with the database, new insights on diaphasic variation in
Agricola’s works will hopefully appear in the future.
Recently, a suggestion has been made concerning the role of printers,
typesetters and typography in general on Agricola’s spelling (see Perälä
2007: 10–40). Agricola’s works were printed in Stockholm, since the first

360 Taru Nordlund
printing house in Finland was opened only in 1642. However, Agricola had
a personal representative in Stockholm, who supervised the printing, and he
himself also often visited the printing house. Therefore, it has been as-
sumed that the correctors of the printing house – who probably did not
know Finnish at all – did not play a significant role in his spelling (Hein-
inen 2007: 158–163, Perälä 2007: 29). Perälä (2007: 20) has pointed out
that the typography in Agricola’s works closely resembles the works of the
printing houses in Northern and Central Germany, and especially the books
of Luther printed in Saxony.
Some variation in the early spelling could be explained by the general
practices of the printing houses (Perälä 2007: 29). To cut down expenses,
16th-century works were printed in a dense and compact form. The printing
surface was ideally very homogeneous and the margins straight. The print-
ers did not want to use wide spaces between words to attain this. Rather, it
is possible that the lines were made straight at the expense of orthography:
a large number of sounds in several words could be written with either one
or two letters, for example hedelmä / hedhelme ‘fruit’, perkele / perchele
‘Satan’. In addition, abbreviations and typographic ligatures (e.g., æ) were
also used. It is noteworthy that the variation between short and long sounds
could also have provided printers with options to treat the layout of the
page. A typographical approach of this kind is certainly in resonance with
current views on the variation in vernacular texts from the late Middle Ages
and the Early Modern period elsewhere in Europe (see for example Voeste
2008).
5. The development of orthography in the successive editions
of the Bible
Quite a lot of the orthographical variation seen in the texts of Agricola and
his contemporaries was regularized in the first translation of the Bible into
Finnish in 1642. From this time onwards, the standardizers apparently
aimed at a one-to-one correspondence between graphemes and phonemes.
In Agricola’s writings, there were still many cases where a single grapheme
had several functions. In the first edition of the Bible, the graphemes <q>
and <ki> disappeared for the notation of /k/. However, the distinction be-
tween /k/ preceding a front vowel (noted with <k>) and that preceding a
back vowel (noted with <c>) subsisted until the end of the 18th century.
The spelling of long vowels in final syllables remained unstable until the

Finnish 361
19th century (Häkkinen 1994: 179, Lehikoinen and Kiuru 2001: 72, Rapola
1965: 73–76).
Until the 19th century, successive editions of the Bible served as an
authority for the standardization of orthography. Each committee or editor
appointed to this work made slight improvements in the orthography. In
practice, variation decreased and phoneme-grapheme correspondence be-
came the norm. Towards the turn of the 19th century, opposition to “for-
eign” letters increased, for example <g> used for [k] after nasal or liquid
consonants as in hengi (today henki) ‘spirit’, <c> or <x> used for [ks] as in
caxi (today kaksi) ‘two’. This anticipated the beginning of the nationalistic
aspirations, and led to lively debates on Finnish orthography in the first
decades of the 19th century.
Towards the end of the 18th century, and especially along with the Bible
translation of 1776 (the so-called Old Church Bible), orthographical inno-
vations were usually to be found in profane literature. Religious writings
fell behind, and the Bible, which in the early stages of standardization used
to function as a precursor for all innovations, now became archaic, too “sa-
cred” to be manipulated at all. However, this applied to syntax and vocabu-
lary more than to orthography, phonology or morphology. This was the
situation up until 1992, which saw the latest translation of the Bible into
Modern Finnish. This change in attitudes also reflects a change in transla-
tion strategies, as the strategy of faithful translation gave way to more func-
tional approaches.
6. The 19th century: Nation-building reflected
in the standardization of orthography
6.1. Introduction to 19th-century Finland
As mentioned above, literary Finnish was at first mainly based on the west-
ern dialects. However, along with the political and ideological climate of
the 19th century, the dialectal basis of Standard Finnish became wider and
more democratic, as eastern dialects started to have their impact as well. In
the 19th century, the official status of Finnish changed; in 1863 Finnish
was decreed to have equal status with Swedish, and towards the turn of the
century – slowly but gradually – it became a fully-fledged cultural language
that was used, for example, in education, administration, culture, science
and literature.

362 Taru Nordlund
The nationalist movement had a variety of linguistic effects. Old (Stan-
dard) Finnish8 was greatly influenced by Swedish, and 19th-century schol-
ars tried to purify Finnish by ridding it of Swedish loanwords and gram-
matical structures borrowed from Swedish. On the orthographical level, the
phoneme /d/ was attacked on the same basis. In the 1810s to 1820s, there
was heated debate on the phoneme /d/ and its orthography (see, for exam-
ple, Mielikäinen 1996). The most radical writers wanted to abandon /d/
altogether and replace it with its dialect variants. In original Finnish words,
/d/ only appears in word-medial position, as a weak grade form of /t/ in
consonant gradation. In dialects, /d/ is not used,9 and there are several vari-
ants of the weak grade form:
Table 6. Diatopic weak-grade forms of /t/
strong grade weak grade
pata ‘pot’
padan (Standard Finnish) ‘pot-GEN’
paran/palan (western dialects) ‘pot-GEN’
paØan (eastern dialects) ‘pot-GEN’
Despite the debate at the beginning of the 19th century, the phoneme /d/, as
well as the grapheme <d>, remained in Standard Finnish. What were the
reasons for this?
Firstly, abandoning the grapheme <d> and using its dialect variants
would have caused variation in the writing, as there were several different
variants in different dialects. At the same time, the dialect basis of Standard
Finnish was under discussion: some writers wanted to increase the eastern
elements in the standard. Therefore, there was no consensus on which dia-
lect form would be chosen – in fact, in this debate, nobody even suggested
that only one dialect variant could be chosen for the standard.
Secondly, and more importantly, <d> remained in the orthography, as it
had already become a marker of “civilized speech”. As mentioned before,
Swedish had been the language of the educated classes and the intelligent-
sia, and only towards the end of the century did Finnish gradually become
the official cultural language. During the standardization process, many
Swedish-speaking people chose to speak Finnish, and it also became possi-
ble to have a higher education in Finnish. Even though the national-
romantic ideology of the time highly valued Finnish dialects and rural liv-
ing, the language of the Finnish-speaking peasantry as such could not have
served as a model for the language of the educated classes. Thus, the dialect

Finnish 363
variants of <d> were not accepted by the educated classes, being too “pro-
vincial”. And as there originally was no standard educated spoken Finnish
– the language being only spoken by the common people – even the ideal
of the proper way to speak came to be very close to written Finnish. In a
way, spoken Standard Finnish was carefully constructed and artificial, not
based on the dialect of a politically and culturally influential region, as is
the case for the spoken standard in many other European languages (e.g.,
London English or Parisian French). This has affected – and promoted –
strictly normative attitudes on both written and spoken Finnish up to the
present day (Mäntynen 2003: 32–39, Paunonen 2006: 44–47).
6.2. 19th-century attempts to standardize consonant reduplication:
An example of etymological spelling
I will next look at attempts to standardize the notation of the so called con-
sonant reduplication (see section 2) into Finnish orthography: that is, at-
tempts to create a system that would represent this phenomenon in writing.
This is an example of an attempt to create etymological spelling: the ex-
plicit marking of consonant reduplication in writing was thought to reveal
an older and more original stage of the language.
As explained in section 2, word-final /k/ or /h/ was earlier used in many
morphological categories. In final position, before a pause, it was lost
(*veneh > vene ‘a boat’). In the western dialects, the final consonant was
lost earlier than in the eastern dialects, probably before the 14th century,
well before Agricola’s texts. In some eastern dialects, the final /k/ or /h/ is
still heard in final position. In the word-boundary position, between two
words, the final /k/ or /h/ was assimilated with the following consonant, and
before a vowel, it was pronounced as a glottal stop (cf. table 7).
In the earliest known texts, the morphological categories that ended with
a /k/ or /h/ normally end in a vowel. There are sporadic occurrences of con-
sonants in some words (e.g., pereh ‘family’, mod. Finnish perhe) and for
example in the allative case (talollen10 talo-ALLAT, ‘to the house’). The
assimilated form only appeared in writing before clitic particles:
(4) pojallekkin
pojalle
boy-ALLAT
‘to a/the boy as well’
+
+
kin
a clitic particle

364 Taru Nordlund
Table 7. Word-final /k/ and /h/ in different phonological environments
example
position
result of
sound change
[tuleeko veneØ]
come
-IND.PRES.3.S.-
QUES. boat
‘Does the boat come?’
final, before pause
loss
[venet tulee]
boat
come-IND.PRES. 3.S.
‘a/the boat comes’
before a consonant
(between two words)
assimilation
[venev_vaja]
‘boat shed’
before a consonant
(in compound words)
assimilation
[veneʔ ʔehtii]
boat
arrive in time-
IND.PRES.3.S.
‘a/the boat arrives in time’
before a vowel
glottal stop
Thus, in the earliest texts, consonant reduplication is only occasionally
marked. If marked, it only appears within a single word, that is, assimila-
tion is never taken into account in writing if it appears between two sepa-
rate words (Karemo 1971: 10–67).
The earliest grammars11 did not mention consonant reduplication at all.
It was not until the 19th century that this phenomenon started to arouse
interest on the whole. During the first half of the 19th century, several dif-
ferent proposals were made to create a notation for consonant reduplication
in Finnish orthography: for example, the letters <h> and <c>, as well as
different types of diacritics were suggested for this purpose (Karemo 1971:
68–119):
Table 8. Suggestions for the notation of consonant reduplication
suggested form author
sanoh
sanoc
sanok
sano’
sano’
say-IMP.2.S.
‘Say!’
Renvall (1826)
Lönnrot (1841)
Gottlund, e.g., (1829)
von Becker (1824)
Varelius (1845)

Finnish 365
In the mid-1840s, a debate about consonant reduplication and its orthogra-
phy flared up. In his textbook on natural sciences, Enon opetuksia luonnon
asioista [The uncle’s teachings about matters of nature] (1845), Antero Varelius
used the sign <’> after a word to indicate consonant reduplication. As the
following page of his book shows, consonant reduplication was – and still
is – a frequent phenomenon in Finnish:
Figure 1. A page of Varelius’s book Enon opetuksia luonnon asioista (1845: 5)
Varelius used his notation very systematically, both in final position (be-
fore a pause) and between separate lexemes. As was to be expected, Vare-
lius’s text provoked criticism from other writers, which resulted in heated
discussion in newspapers. Varelius used the notation in his own works until
1851. He also received some support from the newspaper Suometar, a pro-
Finnish newspaper founded by Varelius and some of his friends in 1847.
However, Varelius’s proposition did not catch on, and he finally gave it up
himself (Kaasalainen 1988, Karemo 1971: 120–126).

366 Taru Nordlund
Why was the notation suggested by Varelius not accepted? Firstly, there
was (and still is) a lot of variation in the distribution of consonant redupli-
cation in different dialects. Some dialect speakers recognized the system,
but for others, consonant reduplication might have been less of a recog-
nized phenomenon. Even today, consonant reduplication is stronger in
some morphological categories and phonological environments, and weaker
in others. There is regional variation and even idiolectal variation. It is easy
to see that the notation would have been difficult and complicated for many
speakers of Finnish, especially people with no linguistic training to help to
identify the relevant categories. Indeed, some opponents pleaded the igno-
rance of the “peasantry”: it would be unnecessary to complicate texts that
were mainly directed at uneducated people, as Varelius’s textbook was.
Secondly, from the early texts on, the final consonant may have ap-
peared sporadically in writing in certain words, in the allative, and before
the clitic particles, that is, within a single word. However, no sign of the
marking of the geminate between two separate words can be seen (Karemo
1971).12 Even those 19th-century writers who had no formal education and
who were minimally exposed to written language do not have a trace of this
phenomenon in their manuscripts. Some self-educated eastern writers do
have single lexemes that end in -/k/ or -/h/, or words with clitic particles
written with a geminate, but there is no sign of the geminate between two
separate words. Most speakers of Finnish today probably do not recognize
consonant reduplication in their speech. It is subconscious. Either it did not
receive a notation in orthography because it is subconscious, or, it is sub-
conscious because it is not represented in writing. Whatever the explana-
tion, the independence of the single word, at least in writing, was too strong
for a phenomenon that appears between two words to become a part of
standard orthography (Nordlund 2007).
Last, but not least, it was pointed out that the notation Varelius sug-
gested would be “messy” and difficult for printing houses. Printing would
be slow and prone to misprints. Suitable printing letters did not always
exist. For example, as can be seen in figure 1, the sign that came to be used
in Varelius’s own book bore more of a resemblance to the number 6 <6>
than to the apostrophe <’> that it was meant to be.
The final point to make is that, despite the obvious difficulties, why was
it so important for some 19th-century writers to design notations for conso-
nant reduplication? The answer probably lies in the attitude towards the
Finnish language in general. The norms of Standard Finnish were con-
sciously built in the 19th century, and the national-romanticists readily saw
Finnish as a genuine, pure, original, and a beautifully symmetrical lan-

Finnish 367
guage. The marking of the final /k/ and /h/ would in a way represent the
older and fuller form of the language. The older stage of the language
would be systematically reconstructed if the elements that had been lost
were represented by a specific sign in the writing. This was a scientific
attitude towards language, and it reflected the growing interest in the study
of Finnish. More than orthography, however, 19th-century debates were
usually concerned with morphology. There were debates such as: which
ending should be chosen for the inessive case in Standard Finnish? Is the
Finnish negation verb a real verb with full inflection, or a particle that is not
conjugated (see Laitinen 2004)? Thus, for Varelius and others who sug-
gested notations for consonant reduplication, it was probably less important
to have a one-to-one correspondence between pronunciation and writing
than to preserve a morpho-phonological feature of older times (Laitinen
2004, Nordlund 2004, Nordlund 2006).
7. Conclusion
The role of the Western Church was essential in the initial stages of the
development of literary Finnish. Some traces of religious uses of Finnish
can be seen in Catholic times but it was the Reformation that started the
history of written Finnish properly, with the idea that everyone should have
a chance to familiarize himself or herself with the religious message in his
or her own language. The earliest writings in Finnish show a lot of varia-
tion in spelling. This is partly explained by the foreign models of writing
that did not always do justice to the phonological system of Finnish. Dia-
topic variation is also usual, and due to the political and cultural situation in
late medieval Finland, especially features of south-western dialects appear.
Very rarely-used south-western features were removed from the spelling
along with the first edition of the whole Bible in 1642, and during the
19th century, the dialect basis of Standard Finnish became broader as sev-
eral eastern features were accepted for the standard.
An electronic morpho-syntactic database of Mikael Agricola’s works
has just been prepared, and in 2007, the first critical edition of his works
was published (Häkkinen 2007). These projects will hopefully lead to thor-
ough research on the variation seen in Agricola’s spelling, especially the
diaphasic variation in his own works that is for the most part unexplained.13
No doubt, the morpho-syntactic database will also open several new van-
tage points on variation, for example, the role of typography and printing
on the spelling and the process of writing in late medieval Finland.

368 Taru Nordlund
Notes
1. An early variant of the Pater Noster in Finnish is found in the Cosmography of
Sebastian Münster, printed in 1544. This variant includes features of eastern
and Bothnian dialects, and supposedly dates from the Catholic period (Häkki-
nen 1994: 80).
2. A conventional view is that there are 21 phonemes in Finnish: 8 vowels and
13 consonants. In Standard Finnish, all the vowels and most of the consonants
have long variants, that is, double vowels or double consonants, geminates. The
sounds [b], [f], [g] and [ʃ] do not appear in original Finnish words, but only in
loan-words.
3. The phoneme /æ/ is an open front unrounded vowel (as in the English word
hat) and the phoneme /ø/ a close-mid front rounded vowel (as in the French
word deux ‘two’).
4. The following notation is used: the notation /k/ refers to the phonemic level, [k]
reflects pronunciation, the phonetic level, and the notation <k> is used to indi-
cate the graphemic level.
5. As shown in this example, the phoneme /n/ is always pronounced homorgani-
cally with the following velar or bilabial stop. In this case, /ŋ/ does not have a
status as an independent phoneme, but it only functions as a phonetic variant of
/n/ in certain environments.
6. The <ʔ> is the sign of a glottal stop.
7. In addition to the shortening of the vowel, there are other phonological changes
in this phrase as well. These are also typical for south-western dialects.
8. The conventional way to periodize the history of Standard Finnish is as fol-
lows:
– before 1540 Early Finnish
– 1540–1810/1820 Old Finnish
– 1810/1820–1870/1880 Early Modern Finnish and
– from 1870/1880 onwards Modern Finnish.
9. Apart from a small area on the south-western coast where it is considered to be
a language-contact feature from Swedish.
10. The allative case ended either in a phoneme /k/ or a phoneme /n/. The allative
with the ending -llen was used in Finnish texts until the 20th century, however,
always along with the allative that ended in a vowel, which is the present day
norm.
11. The earliest grammars of Finnish date from the 17th century. At first, grammars
were mainly directed towards the Swedish-speaking authorities and officials in
Finland. Descriptive grammars and grammar books for the instruction of Fin-
nish as a mother tongue only appeared in the 19th century.
12. There are at least two obvious exceptions to this. The 18th-century lexicogra-
pher, folklorist and writer, Christfrid Ganander, used the final -k in his collec-
tions of folklore, as well as the 19th-century scholar and writer Carl Axel
Gottlund, who aimed to use some kind of eastern Finnish dialect in his writings.

Finnish 369
13. Suggestions have been made that the variation could partly be explained by the
fact that there were several authors working on the translation of the Bible,
along with Agricola. However, no systematic study has been made of this so
far.
References
Primary sources
Becker, Reinhold von
1824 Finsk Grammatik. Åbo: Bibel-sällskapets tryckeri.
Biblia
1642 Biblia, Se on: Coco Pyhä Raamattu, Suomexi 1642. [Biblia, i.e. the
complete Holy Scriptures in Finnish 1642.] Facsimile. Porvoo:
WSOY 1986.
Biblia
1776 Biblia, se on: koko Pyhä Raamattu, suomexi alku-raamattuin hebre-
an ja grekan jälkeen wastauudesta ojettu: esipuhetten, lukuin
sisällepitoin, yhtäpitäwäisten Raamatun paikkain osotuxen, ja
lisättyin registerein kanssa 1776. [Biblia, i.e. the complete Holy
Scriptures, translated into Finnish from the Hebrew and Greek
original versions of the Bible. With a foreword, marginalia, con-
cordance and index 1776]. Turku: Frenckell.
Gottlund, Carl Axel
1829 Otava eli suomalaisia huvituksia, osa 1 [Ursa Major or Finnish di-
versions, part 1]. Tukholma [Stockholm]: Norstedt.
Häkkinen, Kaisa (ed.)
2007 Mikael Agricola. Abckiria. Kriittinen editio. [Mikael Agricola’s
ABC-book, a critical edition]. Helsinki: Suomalaisen Kirjallisuuden
Seura.
Lönnrot, Elias
1841 Bidrag till Finska Språkets Grammatik. Suomi 1 (4): 11–39. Helsin-
ki: Suomalaisen Kirjallisuuden Seura.
Münster, Sebastian
1544 Cosmographia. Beschreibung aller Lender. Basel: Heinrich Petri.
URN of the edition (1545): nbn:de:hbz:061:1-4685
Decription of Finland: http://s1.doria.fi/helmi/bk/rv/fem970026/)
Renvall, Gustaf
1826 Suomalainen sana-kirja = Lexicon linguae Finnicae, cum interpre-
tatione duplici, copiosiore Latina, breviore Germanica. Åbo:
Frenckell.

370 Taru Nordlund
Varelius, Antero
1845 Enon opetuksia luonnon asioista [The uncle’s teachings about
matters of nature]. Helsinki: J. Simelüksen perillisten tykönä.
(http://ia700404.us.archive.org/13/items/wrlennp/wrlennp.pdf)
Secondary sources
Hakulinen, Lauri
2000 Suomen kielen rakenne ja kehitys. [The structure and evolution of the
Finnish language.] Helsinki: Helsingin yliopiston suomen kielen lai-
tos. Original edition 1978.
Heininen, Simo
2007 Mikael Agricola. Elämä ja teokset. [Mikael Agricola: Life and writ-
ings.] Helsinki: Edita.
Häkkinen, Kaisa
1994 Agricolasta nykykieleen. Suomen kirjakielen historia. [From Agri-
cola to Modern Finnish. The history of written Finnish.] Helsinki:
WSOY.
Kaasalainen, Erkki
1988 Antero Varelius. Kirjailija – Pappi – Suomalaisuusmies. [Antero
Varelius, writer, minister and a fennoman.] Vammala: Tyrvään
sanomat.
Karemo, Kaarina
1971 Suomen kielen aspiraatio- eli loppukahdennusilmiön säännöttäminen
ja ortografiset ilmenemismuodot Agricolasta nykyaikaan. [Conso-
nantal reduplication from Agricola to Modern Finnish.] Unprinted
thesis, Department of Finnish, University of Helsinki.
Karlsson, Fred
1983 Finnish Grammar. Helsinki: WSOY.
Laaksonen, Kaino, and Anneli Lieko
2003 Suomen kielen äänne- ja muoto-oppi. [Finnish phonology and mor-
phology.] Helsinki: Finnlectura.
Laitinen, Lea
2004 Kieltosana ja kieletär. Yhden kielikiistan kulku ja ideologiat. [Nega-
tion verb: A case study on language debates and language ideolo-
gies.] In Yhteistä kieltä tekemässä. Näkökulmia suomen kirjakielen
kehitykseen 1800-luvulla, Katja Huumo, Lea Laitinen, and Outi Pa-
loposki (eds.), 177–222. Helsinki: Suomalaisen Kirjallisuuden Seura.
Lehikoinen, Laila, and Silva Kiuru
2001 Kirjasuomen kehitys. [Evolution of written Finnish.] Uudistettu lai-
tos. Helsinki: Helsingin yliopiston suomen kielen laitos.

Finnish 371
Maliniemi, Aarno
1955 Kansankieli Suomen keskiajan kirkossa. [The vernacular in the Fin-
nish church in medieval times.] Novella Plantatio: 82–112. Suomen
Kirkkohistoriallisen Seuran toimituksia 56. Forssa: Forssan kirja-
paino.
Mäntynen, Anne
2003 Miten kielestä kerrotaan. Kielijuttujen retoriikkaa. [Talking about
language. The rhetoric of language columns.] Helsinki: Suomalaisen
Kirjallisuuden Seura.
Mielikäinen, Aila
1996 Reinhold von Becker ja murteiden taistelu. [Reinhold von Becker
and the debate on dialects.] In Kuuskymppinen, Seppo Pekkola (ed.),
126–156. Jyväskylä: Jyväskylän yliopiston suomen kielen laitos.
Nordlund, Taru
2004 Arat taimet ja ankarat puutarhurit  1800-luvun lopun kielikeskuste-
lun metaforat ja kieli-ideologiat. [Metaphors and ideologies of lan-
guage debates in the latter part of the 19th century.] In Yhteistä kieltä
tekemässä. Näkökulmia suomen kirjakielen kehitykseen 1800-luvulla,
Katja Huumo, Lea Laitinen, and Outi Paloposki (eds.), 286–322.
Helsinki: Suomalaisen Kirjallisuuden Seura.
Nordlund, Taru
2006 Kielioppia rakentamassa: otta- ja oitta-verbien normittamisen his-
toriaa. [Building the grammar: history of standardization of o(i)tta-
verbs.] In Kielioppi koulussa. Äidinkielen opettajain liiton vuosikirja
XLX, Minna Harmanen, and Mari Siiroinen (eds.), 55–69. Helsinki:
Äidinkielen opettajain liitto.
Nordlund, Taru
2007 Double diglossia – lower class writing in 19th-century Finland. Mul-
tilingua 26 (2/3): 229–246.
Paunonen, Heikki
2006 Vähemmistökielestä varioivaksi valtakieleksi. [From a minority lan-
guage to a varied dominant language.] Helsinki: Helsingin yliopiston
suomen kielen ja kotimaisen kirjallisuuden laitos.
Penttilä, Aarni
1932 Upsalan suomenkielisen (1500-luvulta polveutuvan) evankeliumikir-
jan fragmentin kielestä. [On The language of Upsala Gospel Book
from the 16th century.] Suomi 5 (13), 1–138.
Perälä, Anna
2007 Mikael Agricolan teosten painoasu ja kuvitus. [Typography and
illustration in Mikael Agricola’s works.] In Mikael Agricola Ab-
ckiria. Kriittinen editio, Kaisa Häkkinen (ed.), 10–40. Helsinki:
Suomalaisen Kirjallisuuden Seura.

372 Taru Nordlund
Rapola, Martti
1965 Suomen kirjakielen historia pääpiirteittäin. I Vanhan kirjasuomen
kirjoitus- ja äänneasun kehitys. [History of Old Biblical Finnish, Or-
thography and phonology.] Helsinki: Suomalaisen Kirjallisuuden
Seura.
Sulkala, Helena, and Merja Karjalainen
1992 Finnish. (Descriptive Grammars) London/New York: Routledge.
Voeste, Anja
2008 Orthographie und Innovation. Die Segmentierung des Wortes im
16. Jahrhundert. Hildesheim/Zurich/New York: Olms.