Instant Download Yanqi Lake Lectures On Algebra Wen-Wei Li PDF All Chapters
Instant Download Yanqi Lake Lectures On Algebra Wen-Wei Li PDF All Chapters
com
https://textbookfull.com/product/yanqi-lake-
lectures-on-algebra-wen-wei-li/
https://textbookfull.com/product/proceedings-of-the-6th-conference-on-
sound-and-music-technology-csmt-revised-selected-papers-wei-li/
textbookfull.com
https://textbookfull.com/product/thermal-energy-storage-analyses-and-
designs-pei-wen-li-and-cho-lik-chan-auth/
textbookfull.com
https://textbookfull.com/product/lectures-on-geometry-woodhouse/
textbookfull.com
https://textbookfull.com/product/21-kesaris-the-untold-story-of-the-
battle-of-saragarhi-1st-edition-kiran-nirvan/
textbookfull.com
https://textbookfull.com/product/health-protection-principles-and-
practice-1st-edition-samuel-ghebrehewet/
textbookfull.com
Shine your Icy Crown (You Are Your Own Fairy Tale #2) 1st
Edition Amanda Lovelace
https://textbookfull.com/product/shine-your-icy-crown-you-are-your-
own-fairy-tale-2-1st-edition-amanda-lovelace/
textbookfull.com
Counselling and Medical Decision Making in the Era of
Personalised Medicine A Practice Oriented Guide 1st
Edition Giovanni Boniolo
https://textbookfull.com/product/counselling-and-medical-decision-
making-in-the-era-of-personalised-medicine-a-practice-oriented-
guide-1st-edition-giovanni-boniolo/
textbookfull.com
YANQI LAKE
LECTURES ON
ALGEBRA
/ Wen-Wei Li
• Exercise 1.4.2 Need an extra condition 𝑝 > 2, otherwise there are obvious coun-
terexamples.
• Theorem 3.5.6 There are gaps in this proof. Please refer to J. Neukirch, Algebraic
Number Theory, Chapter IV. §3 (especially (3.3) Theorem) for a complete proof
which involves the use of Pontryagin duality, etc.
• Section 8.2 The exposition here is in chaos. Please see the Part 3, §3.2 of these
lecture notes for an improved version.
• Lemma 12.2.3 The definition of ind𝐺 𝐻 (𝑊) → 𝑃(𝑊) should map 𝑓 to ∑𝑔∈𝐻\𝐺
̄ 𝑔−1 ⊗
𝑓 (𝑔), where 𝑔 is any representative of the coset 𝑔.̄
The inverse map 𝑃(𝑊) → ind𝐺
𝐻 (𝑊) is obtained as follows: write 𝐹𝐺 ⊗𝐹𝐻 𝑊 =
⨁ 𝑔−1 𝐹𝐻 ⊗𝐹𝐻 𝑊. Given 𝑔 ∈ 𝐺, let 𝑔̄ ∈ 𝐻\𝐺 be the coset it belongs to. Then
𝑔∈𝐻\𝐺
̄
𝑔 furnishes an isomorphism from the summand 𝑔−1 𝐹𝐻 ⊗𝐹𝐻 𝑊 = 𝑔−1 ⊗ 𝑊 to
𝑊, namely by discarding the first tensor slot 𝑔−1 . This allows us to associate to
any element of 𝐹𝐺 ⊗𝐹𝐻 𝑊 a function 𝑓 ∶ 𝐺 → 𝑊, which is readily seen to be in
ind𝐺
𝐻 (𝑊).
1
Yanqi Lake Lectures on Algebra
Part 1
Wen-Wei Li
Chinese Academy of Sciences
Email: wwli@math.ac.cn
Version: 2015-01-14
Foreword vii
Backgrounds 1
1 Field extensions 3
1.1 Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 Algebraicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 The algebraic closure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4 Splitting fields and normality . . . . . . . . . . . . . . . . . . . . . . . . . 9
4 Modules 37
4.1 Review: rings and ideals . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.2 Modules: basic definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
4.3 Direct sums and free modules . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.4 Exact sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.5 Chain conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.6 Hilbert basis theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
iii
iv
7 Semisimple rings 71
7.1 Wedderburn-Artin theory for semisimple rings . . . . . . . . . . . . . . . 71
7.2 Double centralizer property . . . . . . . . . . . . . . . . . . . . . . . . . . 76
7.3 Another approach to the Wedderburn-Artin Theorem . . . . . . . . . . . 77
7.4 Jacobson radicals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
8 Semiprimitive rings 81
8.1 Semiprimitivity versus semisimplicity . . . . . . . . . . . . . . . . . . . . 81
8.2 Intermezzo: der Nullstellensatz . . . . . . . . . . . . . . . . . . . . . . . . 83
8.3 Primitive rings and primitive ideals . . . . . . . . . . . . . . . . . . . . . 84
8.4 Density theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
8.5 Structure theory for primitive rings . . . . . . . . . . . . . . . . . . . . . . 86
8.6 The primitive spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
8.7 Finite-dimensional algebras: Burnside’s Theorem . . . . . . . . . . . . . . 88
Bibliography 163
Index 167
vi
FOREWORD
These lecture notes were prepared for the graduate course Algebra I (210002H) during
September 2014 – January 2015 at the University of Chinese Academy of Sciences, Yanqi
Lake campus. The last few chapters on non-commutative rings and representation the-
ory are based on earlier lectures during 2013-2014. Each lecture, or more appropriately
each chapter in these notes, took roughly one week.
This is not a standalone lecture series on algebra. We presume some knowledge
about:
(i) undergraduate abstract algebra, including the notions of groups and rings;
(ii) rudimentary set theory, namely some familiarity with cardinalities;
(iii) a certain “common sense” about categories and functors — the student is expected
to take some statements for granted.
For the relevant backgrounds we will often refer to [11, 12, 16]; however, there was
no prescribed textbook during our course. The author benefited a lot from [16, 12, 14,
15, 22], as well as other textbooks in Chinese which are not included in the bibliography
due to TEXnical difficulties.
Our course was taught in 90% Taiwanese-accented mandarin and the lecture notes
were delivered weekly. As a result these notes were written rather hastily with lots
of mistakes in both mathematics and English, and they reflect the author’s eccentric
mathematical taste as well. Due to the manner in which these notes were prepared,
there are minor inconsistencies in notations and there are no cross-references between
different chapters. Categories are mentioned but the discussion is far from adequate;
the expositions on representation theory are especially unsatisfactory. The author takes
full responsibility for all these defects.
As for the photos and pictures in these notes (usually irrelevant), their sources are
explicitly stated whenever possible.
These notes are certainly not intended for publication. Nonetheless some coura-
geous people might find them useful. The author would like to express his deep grati-
tude to all the students attending this course for their patience, tolerance and constant
support, as well as many corrections.
◆ The cover page uses the fonts Bebas Neue and League Gothique, both licensed
under the SIL Open Font License.
vii
Visit https://textbookfull.com
now to explore a rich
collection of eBooks, textbook
and enjoy exciting offers!
viii
Wen-Wei Li
Zhongguancun, Beijing,
January 2015
BACKGROUNDS
Throughout this course, the reader is assumed to have acquaintance with undergrad-
uate algebra, namely the basic notions about sets, groups, rings and modules. Details
can be found in any decent textbook such as [11]. In order to recall the relevant notions
(in English!) and to fix the notations, we give a recapitulation below.
Sets We work in the framework of ZFC set theory. The usual operations on sets are:
∩, ∪, ×, ⊔ (= disjoint union); the Cartesian product (resp.Qintersection,Tunion,
disjoint union)
S F of a family of sets {E i : i ∈ I} is denoted by i∈I E i (resp. i∈I E i ,
E
i∈I i , i∈I ); the set of maps from X to Y are denoted by Y X . The cardinality
of a set E is denoted by |E| or #E. For the most part in this course, we neglect
set-theoretic issues such as proper classes, etc.
If f : X → Y is a map and E ⊂ Y, we write f −1 (E ) : {x ∈ X : f ( x ) ∈ E}; when
E { y} we use the shorthand f −1 ( y ) f −1 ( { y} ) , commonly called the fiber of f
over y.
The symbol A : B reads as “A is defined to be B”. The arrow ֒→ (resp. ։) means
an injection (resp. surjection), and x 7→ y means that the element x is mapped
to y. If ∼ is an equivalence relation on a set E, the corresponding quotient set is
denoted by E/ ∼.
We admit Zorn’s Lemma: let ( P, ≤ ) be a partially ordered set. If every chain (i.e.
totally ordered subset) of P has an upper bound in P, then there exists a maximal
element in P. Zorn’s Lemma is known to be equivalent to the Axiom of Choice.
1
2
Rings Unless otherwise specified, the rings are assumed to have multiplicative unit
element 1. Therefore, a ring R is an additive group ( R, +, 0) together with a mul-
tiplication map ( x, y ) 7→ x · y x y that makes ( R, ·, 1) into a monoid. These
structures are related by distributivity:
x ( y + z ) x y + xz, z ( x + y ) zx + z y.
The standard example for a commutative ring is the ring of integers , the non-
commutative case is best illustrated by the ring of n × n-matrices. An element
x ∈ R is called invertible (or a unit of R) if ∃y ∈ R, x y yx 1; in this case y
is unique and we denote it as x −1 . The units form a group under multiplication,
denoted as R × .
Homomorphisms are maps that respect algebraic structures, namely the conditions
such as ϕ ( x y ) ϕ ( x ) ϕ ( y ) and ϕ (1) 1 are imposed. For a homomorphism ϕ between
groups (resp. rings), we denote its kernel and image by ker ( ϕ ) : ϕ−1 (1) (resp. ker ( ϕ ) :
ϕ −1 (0) ) and im ( ϕ ) .
There is also a notion of “substructures”, namely the subgroups, subrings, etc. A
subgroup N ⊂ G is called normal if xN x −1 ⊂ N for all x ∈ G, in which case we write
N ⊳ G. To a normal subgroup one associates the quotient group G/N.
As for rings, it turns out that the two-sided ideals play a rôle similar to that of normal
subgroups. Let R be a ring. An additive subgroup I of R is called a (two-sided) ideal if
xI ⊂ I and Ix ⊂ I for all x ∈ R. For commutative rings one may simply speak of ideals,
without specifying the sides. The quotient ring R/I is the additive quotient group R/I
equipped with the multiplication ( x + I )( y + I ) x y + I.
The two-sided ideal generated by elements x1 , . . . , x n ∈ R will be written as
( x1 , . . . , x n ) .
1.1 Fields
We begin by reviewing the rudiments of field theory. Any ring R admits exactly one
homomorphism from , namely
−→ R
a 7−→ a · 1.
Its image must be of the form /p for a uniquely determined integer p ≥ 0. Assume
that R has no zero-divisors, i.e. x y 0 ⇐⇒ x 0 ∨ y 0, then so is /p , and one
concludes immediately that p is either a prime number or zero.
Definition 1.1.1. Let F be a field. Its characteristic is the number p above. Denote it by
char ( F ) .
Definition 1.1.2. An intersection of subfields of F is still a subfield, thus it makes sense
to talk about the smallest subfield inside F. Call it the prime field of F.
Any subfield must contain 1, 0 and every expression that can be obtained from field-
theoretic operations. Thus
⋆ either char ( F ) 0, in which case ֒→ F and we obtain a copy of inside F by
inverting the nonzero integers;
⋆ or char ( F ) p > 0, in which case we obtain a copy of /p : p inside F.
Summing up, the prime field of F is or p , according to whether char ( F ) is zero or a
prime number p.
Next comes the notion of compositum.
Definition 1.1.3. Let F, F′ be two subfields of an ambient field L. Their compositum,
written as FF′, is the smallest subfield of L containing both F and F′. More concretely,
the elements of FF′ take the form
x1 x 1′ + · · · x n x ′n
′ ∈ L
y1 y1′ + · · · y m y m
3
4
Note that a ring homomorphism between fields ϕ : F → E must have kernel equal
to {0}. Thus, instead of talking about homomorphisms, one may concentrate on embed-
dings of a field F into another field. If E ⊃ F, we say that E is a (field) extension of F; it
is customary to write such an extension as E/F — do not confuse with quotients! Field
extensions will be the main concern of this lecture.
Let E/F be an extension. Note that E forms an F-vector space: the addition in E and
the scalar multiplication of F on E come from their ring structures.
Definition 1.1.4. The degree of E/F is defined as dimF E, also written as [E : F]. Exten-
sions of finite degree are called finite extensions.
[L : F] [L : E][E : F]
as cardinal numbers. In particular, L/F is finite if and only if L/E and E/F are both finite.
Proof. Choose a basis B (resp. C) of the F-vector space E (resp. of the E-vector space L).
Every element v ∈ L has a be uniquely expression
X
v γc c (finite sum), γc ∈ E.
c∈C
P
Expanding each γc as an F-linear combination γc b∈B γb,c b, we arrive at a unique
expression X
v γb,c bc, γb,c ∈ F.
b∈B
c∈C
This provides a basis for L which is in bijection with B × C, proving our assertions.
1.2 Algebraicity
The innocent-looking notion of finiteness is directly related to algebraicity, as reviewed
below. Consider an extension E/F. For any element u ∈ E, we write F ( u ) as the subfield
generated by u, that is: \
F (u ) E′ ⊂ E.
F⊂E′ ⊂E
u∈E′
F[u] : {P ( u ) : P ∈ F[X]}
5
Figure 1.1: Ernst Steinitz (1871-1928) initiated the axiomatic study of fields (in German:
der Körper) in [25]. He also introduced the fundamental concepts such as prime fields,
transcendence degree, etc. (DOI: 10.1515/crll.1910.137.167)
[Q ∈ F[X], Q ( u ) 0] ⇐⇒ P|Q.
F ( X ) −→ F ( u )
Q/R 7−→ Q ( u ) /R ( u ) , Q, R ∈ F ( X ) , R , 0
Note that this is just an abstract result: it is not so easy to determine the minimal
polynomial of α+β, etc. in practice. Another consequence is that the algebraic elements
in E forms a subextension Ealg /F.
√ √
Exercise 1.2.5. Determine the minimal polynomial of the algebraic number 2 + 3
over .
An extension E/F is called algebraic if every element u ∈ E is algebraic over F.
Proposition 1.2.2 implies that E/F is algebraic if and only if it is a union of finite exten-
sions of F.
Exercise 1.2.6. If L/E and E/F are both algebraic, then L/F is algebraic as well.
At the other extreme, given a field F and a possibly infinite set Γ, we may form
the field F (Γ) of rational functions with indeterminates in Γ; when Γ {X1 , . . . , X n }
we recover the familiar F ( X1 , . . . , X n ) . Unlike the algebraic setting, Γ is algebraically
independent: there are no non-trivial polynomial relations among elements in Γ. It can
be shown that every field extension E/F has a decomposition
E ⊃ F (Γ) ⊃ F
algebraic
∼ ∼
F (u ) F[X]/ ( P ) F (v ) E
u X v
in which every arrow is an F-embedding.
8
( ι : E → F̄′ ) ≤ ( ι1 : E1 → F̄′ ) ⇐⇒ [E ⊂ E1 , ι1 |E ι] .
It is easy to see that ( P, ≤ ) is a partially ordered set. We want to apply Zorn’s Lemma to
get a maximal ι : E → F̄′; indeed, every chain in ( P, ≤ ) has an upper bound — simply
take union! By Lemma 1.3.2, maximality implies E F̄.
It remains to show that ι ( F̄ ) F̄′. To see this, note that the algebraically-closeness
of F̄ transports to ι ( F̄ ) . This implies ι ( F̄ ) F̄′, since for every u ∈ F̄′, the roots of the
minimal polynomial of u over F already lie in ι ( F̄ ) .
As for the existence of F̄, one seeks some sort of “maximal algebraic extension” of
F and the construction is again based on Zorn’s Lemma. However, manipulating the
collection (hum?) of all algebraic extensions of F is somehow hazardous. So we appeal
to the following device: there exists a set Ω such that for every algebraic extension E/F,
the set E is in bijection with a subset of Ω. The basic idea is sketched as follows. write
[
E En , E n : {u ∈ E : [F ( u ) : F] n}.
n≥1
For every n, the map that associates u ∈ E n with its minimal polynomial S over F is
at most n-to-1, so everything boils down to bound the cardinaltiy of F[X] n≥1 {P :
deg P n}.
Now we consider the nonempty partially ordered set formed by algebraic exten-
sions E/F, where E ⊂ Ω set-theoretically, and ≤ is defined by field extension. Again,
Zorn’s Lemma implies the existence of some maximal E/F. If E is not algebraically
closed, we may construct an extension E′/E with ∞ > [E′ : E] > 1 by Lemma 1.2.2. The
set E′ being algebraic over E, thus over F by Exercise 1.2.6, it can be re-embedded into
Ω; this would violate the maximality of E.
See [16, p.231] for another famous proof due to E. Artin. It also relies on Zorn’s
Lemma, however.
Lemma 1.3.5. Let K/F be an algebraic extension, then every F-embedding ι : K → K is an
F-automorphism.
Proof. Let v ∈ K and denote its minimal polynomial over F by P. Enumerate the roots
of P inside K as v v1 , . . . , v n and set K0 : F ( v 1 , . . . , v m ) , which is finite over F. It
follows that ι induces an F-embedding K 0 → K 0 , which must be an F-automorphism
for dimensional reasons. As v is arbitrary, the surjectivity follows at once.
9
Conversely, the splitting field of Pu lies in E for every u ∈ E since we have seen that (ii)
implies (i).
(iii) ⇒ (ii): Suppose that E is the splitting field of {Pi :∈ I} inside F̄. Let ι : E → F̄ be
an F-embedding, it suffices to show that ι induces an F-automorphism of the splitting
field of each Pi . This is clear since ι permutes the roots of Pi .
LECTURE 2
SEPARABILITY AND FINITE GALOIS
EXTENSIONS
2.1 Separability
We always fix a ground field F. If E, E′ are two extensions of F, we denote by HomF (E, E′ )
the set of F-embeddings E → E′. Similarly, we define the group of F-automorphisms
AutF (E ) , etc.
Consider a finite extension F ( u ) /F generated by a single element u, and let L/F be
an algebraic extension in which P, the minimal polynomial of u over F, splits into linear
factors. Recall that we have established the bijection
∼
HomF ( F ( u ) , L ) −→ {v ∈ L : P ( v ) 0}
ϕ 7−→ ϕ ( u ) .
In particular, |HomF ( F ( u ) , L ) | ≤ deg P. Strict inequality can hold when the character-
istic p : char ( F ) is positive. This leads to the notion of separability.
Pn
Definition 2.1.1. Let P ∈ F[X], P ( X ) k0 a k X k . Define its derivative formally as
n
X
′
P ( X ) : ka k X k−1 ∈ F[X].
k1
11
12
Notice that the criterion ( P, P ′ ) , 1 can be checked inside the ground field F, say by
the Euclidean division procedure.
Q
Proof. Write P ∈ L[X] as nk1 ( X − a k ) with a1 , . . . , a n ∈ L being the roots. A straight-
forward manipulation gives the first assertion. When P is irreducible, ( P, P ′ ) , 1 ⇒
P|P ′ ⇒ P ′ 0 since deg P ′ < deg P.
Definition 2.1.3. A polynomial P ∈ F[X] is called separable if it has no multiple roots,
i.e. ( P, P ′ ) 1.
P
We turn to the study of irreducible polynomials P ( X ) k a k X k with P ′ 0. This
is equivalent to ka k 0 for all k ≥ 1. When char ( F ) 0, the only candidates are the
constant polynomials. Assume hereafter that
P
Write P P1 ( X p ) by taking P1 ( X ) p|k a k X k/p . If P1′ 0, the procedure can be
iterated so that eventually
m
(2.1) P (X ) P ♭ (X p ) , P ♭ ∈ F[X], ( P ♭ )′ , 0.
for some m ∈ ≥0 . Note that P ♭ is irreducible since P is. Fix an algebraic closure F̄/F.
It turns out that
−m
(2.2) {α ∈ F̄ : P ( α ) 0} {β p : P ♭ ( β ) 0},
−m
where β p is the p m -th root of β in F̄. In fact, we have
m
−m
pm
X p − β X − βp
x x!
over F̄; this is because p · 1 0 in F̄ and the binomial coefficient y ( x−y ) !y! satisfies
!
p
p| , 0 < a < p,
a
hence
Exercise 2.1.4. The study of purely inseparable polynomials can be further reduced to
m
the case b < F p . Under this assumption, show that the polynomial X p −b is irreducible.
Use this to produce examples of inseparable field extensions.
Now we revert to the case of general characteristic and resume the study of embed-
dings.
Definition 2.1.5. Let E/F be an algebraic extension, define its separable degree as [E :
F]s : |HomF (E, F̄ ) |. This is independent of the choice of the algebraic closure F̄/F.
Lemma 2.1.6 (Tower property). For a tower L/E/F of algebraic extensions, we have [L : F]s
[L : E]s [E : F]s as cardinal numbers.
Proof. Extending the inclusion F ֒→ F̄ to τ : L → F̄ is equivalent to (i) extending it
σ
to various σ : E → F̄, and then (ii) extending each E − → σ (E ) ֒→ F̄ to τ : L → F̄.
There are [E : F]s choices for the first step. As regards the second step, since [L : E]s is
independent of the choice of the embedding of E into F̄ Ē, there are [L : E]s choices
for each σ.
Definition-Proposition 2.1.7. Let E/F be a finite extension, then [E : F]s [E : F]. Call
E/F a separable extension if [E : F]s [E : F].
Proof. Choose u 1 , . . . , u n so that E F ( u 1 , . . . , u n ) . Using the tower
(2.4) E F ( u 1 , . . . , u n ) ⊃ F ( u1 , . . . , u n−1 ) ⊃ · · · ⊃ F ( u 1 ) ⊃ F
and the tower properties of [E : F]s and [E : F], we reduce immediately to the case E
m
F ( u ) . Let P ∈ F[X] be the minimal polynomial of u, and express it as P ( X ) P ♭ ( X p ) as
in the earlier discussions, where P ♭ is separable. It follows that deg P [F ( u ) : F] equals
[E : F]s deg P ♭ (which is the number of distinct roots of P) times [E : F]i : p m .
We have just used the observation that F ( u ) /F is separable if and only if u has sep-
arable minimal polynomial. In this case we say u is a separable element. If u ∈ E is
separable over F, then u is separable over any intermediate field between E and F —
indeed, if a polynomial has no multiple roots, then the same holds for its factors.
Lemma 2.1.8. A finite extension E/F is separable if and only if every u ∈ E is separable.
Proof. Consider the tower (2.4). If every u i has separable minimal polynomial over F
(hence over any intermediate field), the tower properties will give [E : F]s [E : F].
Conversely, we may realize any given u ∈ E as the u 1 in (2.4). The tower property, the
hypothesis [E : F]s [E : F] together with the bounds [· · · ]s ≤ [· · · ] imply [F ( u1 ) :
F]s [F ( u1 ) : F], whence the separability of u u 1 .
Hence we may extend the notion of separability to arbitrary algebraic extensions as
follows.
Definition 2.1.9. An algebraic extension E/F is called separable if every element in E
is separable over F.
Exercise 2.1.10. If L/E and E/F are separable, then so is L/F.
14
We say a field L is separably closed if any separable irreducible polynomial has a root
in L. As in the case of algebraic extensions, there is a notion of separable closure F sep /F,
which is a separable extension with F sep separably closed. Again, we have:
⋆ Existence of F sep /F: simply take the subextension of F̄/F generated by all sepa-
rable elements, or the compositum of all separable subextensions of F̄/F. In fact,
this is the only choice of a separable closure sitting inside F̄!
sep sep
⋆ Uniqueness up to F-isomorphisms: let F1 and F2 be two separable closures.
sep
Embed them into algebraic closures, say Fi ⊂ F̄i for i 1, 2. Since there exists an
∼ sep sep
F-isomorphism τ : F̄1 → F̄2 , we reduce immediately to the case that F1 , F2 ⊂ F̄,
sep sep
and it has been observed that F1 F2 in this case.
Proof. We may assume F sep ⊂ F̄. Then it is the splitting field of the family of separable
irreducible polynomials over F.
Definition 2.2.1. Call an algebraic extension E/F purely inseparable if every element u ∈
m
E satisfies u p ∈ F for some m.
∞
We use the shorthand E ⊂ F 1/p for the last condition defining pure inseparability.
∞
It makes
S perfect sense if E is embedded into an algebraic closure F̄ and F 1/p is taken
m
to be m {u ∈ F̄ : u p ∈ F}, which forms a subfield by (2.3).
Note that [E : F]s 1 if E is purely inseparable, since we have observed that a poly-
m
nomial of the form X p − b has only one root in F̄. The assertions below are immediate.
Exercise 2.2.2. If L/E and E/F are purely inseparable, then so is L/F. A compositum of
purely inseparable extensions of F is still purely inseparable.
Proposition 2.2.3. Suppose E/F is an algebraic extension with p : char ( F ) > 0. Let E s be
the maximal separable subextension, which makes sense by the preceding exercises. Then E/E s
is purely inseparable. When E/F is finite, we have [E : F]s [E s : F] and [E : F]i [E : E s ].
Random documents with unrelated
content Scribd suggests to you:
olvasták kegyelem utján: 5 év. (Minthogy a két évi vizsgálati fogság
nem volt betudva, összesen 7 évi fogság lebegett fejem fölött –
ebből 5 évet szerencsésen ki is töltöttem), mikor vége volt,
visszamentem társaim közé; nyakamba borultak, kezem
szorongatták, s az az osztrák tábornok hozzám lépve kezet adott, s
igy szólt: „Derék fiatal ember ön, kitartást kivánok nehéz sorsában!“
Vége lévén a szomoru czeremóniának mégis némi könnyebbülést
éreztünk a bizonyosság által; s részint az emberi természet folytán,
részint a magyar optimismus közreműködése következtében
egyetlen egy se volt, ki azt hitte volna, hogy azt az időt a fogságban
kitöltjük. Forradalom, háboru, vagy tartós békéből egyiránt
szabadulásunkat combináltuk.
Annyi bizonyos, hogy vigan ebédeltünk, és csak estefelé romlott
el kedvünk, mikor lehozták a sok vasat. Egy karikát az egyik lábra
tettek, a mást a keresztbe álló kézre kellett volna tenni, de elnézés
folytán többnyire balláb és balkézre tették, s a két karikát láncz köté
össze 3, 4, 5 fontos volt az egész szerkezet. Ez eleinte borzasztónak
tetszett, mert szabályszerüleg csak két óráig vehettük le reggel egy
és este egy óráig; azaz: öltözés és vetkezés idejére – és vele kellett
volna hálni is – de már Fehérvárt privát kulcsokat szereztünk a
lakathoz, s a következő három év alatt összevéve nem aludtam
vasban többször, mint 20–30 éjjel. Aztán később kivált Józsefstadban
nappal se igen viseltük. Erről később.
Az itélet kimondását követő hat hét volt fogságunk
legérzékenyebb ideje. Megengedték, hogy rokokinkat értesitsük
itéleteinkről s megengedték, hogy ők hozzánk jöhessenek bucsuzni.
Én szüleimhez a következő levelet intéztem:
Fehérvár, 1854. márcz. 17-ik. Kedves szüleim!
Megragadom az igen tisztelt előljáróság által engedett alkalmat s
irok önöknek. És már most bizonyosan meg is irom, mi sors vár
reám. Az igen tisztelt törvényszék a katonai törvények nyomán, 12
évi várfogsággal büntetni elhatározta, de legmagasabb kegyelme Ő
Felségének, azt 5 évi várfogságra szállitotta. Ez elég szomoru ugyan
mind magokra, mind különösen rám nézve – de én oly erőt kivánok
magoknak, a milyen nekem van. Vigasztalnom fiui kötelességem, de
el nem titkolhatom a fájdalmat, mit érzek ha magokra, testvéreimre,
gyermek örömeim lefolyt óráira s ifjuságom könyelmüen eljátszott
szép napjaira gondolok!… Nem bucsuzom, az elválást jelentő szavak
fájdalmat okoznak. Legyenek nyugodtak, tegyék meg a mit lehet,
vessék reményöket istenbe s ő cs. k. apostoli felsége
kegyelmességébe, s ha ezek nem segitenek, csak erőt kivánok a
várakozásra, mig én megjövök, hogy mindezt elfeledtessem… Nem
sokára elszállitanak rendeltetésünk helyére, hogy hová még most
nem tudom, de ha szabad lesz, megirom; a német és franczia
könyveket várom.
Édesapámnak s édesanyámnak kezeit s testvéreimet ezerszer
csókolva maradok.
Édes szüleimnek
engedelmes fiok.
A többiek is hasonló leveleket irva, Erdély minden részéből
megindultak a szomoru rokonok; a szülők, a nők, testvérek és
gyermekek – Kolozsvárról, Vásárhelyről, a megyékből s különösen a
székely székekből, mind elvándoroltak, hogy a szomoru látogatást
megtegyék. Engedték, hogy három, négy napot a városban
töltsenek, s a délelőttet és délutánt együtt beszélgessék át a profósz
lakásán.
Voltak, kik szivszakgató zokogás nélkül a viszontlátást se tudták
elhordozni, annyival kevésbé az elbucsuzást oly hosszu,
reménytelen, nehéz időre.
Engemet néhai édesatyám, édesanyám és Eszter hugom
látogatott meg; ruhát, fehérnemüt és pénzt hoztak. Mikor
találkoztunk csak édesanyám sirdogált s ugy találták, hogy elég jól
néztem ki, (jó hogy Szebenben nem láttak, de a fehérvári fogságban
a nap nagyobb részét szabadlevegőn töltöttük, kedélyünk derültebb
s kosztunk egészen jó volt). Szüleim három napig voltak ott, s
találkozásainkon nyugodtan beszélgettünk, jó édesatyám
szivarozgatott – s mint általában kevésbeszédü ember csak ennyit
mondott az első nap: „Sok szomoruságot okoztál nekünk édes fiam,
különösen szegény édesanyádnak, de jól esik nekünk, hogy férfiasan
viselted magadat és tudsz türni; elfogatásod után nehány nappal két
ur jött hozzám, kik azt mondák, hogy mondjam meg neked, hogy
mit és miként vallj, és hogy róluk hallgass. Én erre nem válalkoztam,
mert ügyedet nem ismertem, és biztam benned – a mint nekik is
megmondottam – hogy habár nagyon ifju vagy még, de a mihez az
én tanácsom nélkül kezdettél, azt nálam nélkül el is tudod végezni, –
most örömmel látom, hogy nem vagy elcsüggedve s remélem, hogy
ha a poharat fenékig kell is üritened, nem fogok csalódni benned. Az
isten jó, édesfiam s a becsületes embereket soha se hagyja el.“
Ez volt minden, mit édesatyám ez ügyről valaha beszélt; ugy
látszott, hogy ha az ilyes dologban lehetne tanácsot kérni, ő nem
tanácsolta volna a részvétet – de aztán meg volt elégedve
magamviseletével.
Harmadik nap délben bucsuztunk el, édesanyám erősen sirt és
zokogott, kis hugom is folytonosan könnyezett – atyám hallgatva
szivarozgatott. A szomszéd szobákban ifju házaspárok bucsuzódtak
sok esztendőre egymástól; meg egy fogoly apa, és hazatérő fiu –
mondhatom rettenetes perczek voltak – de én bár komoly voltam
egyetlen köny nem jött szememre – csak folytonos beszéllésem
jelezheté, hogy némileg én is meg valék indulva – végre válni kellett,
anyám karjaiból alig tudtam menekülni, kis hugom is nyakamon
csüggött, s midőn édesapám megcsókolt, és kezemet megrázta,
éreztem, hogy egész testében remeg; az én ártatlan lelkü s nemes
szivü jó atyám!…
Talán sokáig én se tudtam volna kitartani – de midőn körültem az
én legkedvesebb szüléim, hugom, s annyi barátom, ismerősöm oly
igaz, oly legyőzhetetlen fájdalommal telve sirtak, zokogtak, – haj!
eszembe jutott a nagy gyász, mely akkor egész hazánkat elboritá –
lelkemben egybe forrt ez egyes emberek busulása a haza nagy
siralmával, s ugyan ki vehetné rosz néven, ha egy pillanatig hazám
szent ügyének diadalán rajongva a hitvallók rémes gyönyörét
éreztem, s bár iszonyu fájdalom dulta keblemet, sirni nem tudtam s
emelt fővel távoztam a bucsuról?!
A bucsulátogatások végre gyérülni kezdtek, mindenikünk
rendezte ügyeit, a hogy tudta, legkeservesebb volt az ifju házasok
állapota, kiknek egy két évi együttlét után 10–12 évre kelle válniok,
épp az élet legszebb korában – a mi az erdélyi nőket illeti,
nagylelküen viselték magukat, és csupán két asszonyról tudom, hogy
más viszonyt kezdett, s máshoz is ment nőül, mig férje a rabságban
szenvedett.
De e két gyenge nőt bőven ellensulyozzák oly valóban meglepő
jellemek, mint Filep Samu, Tompa Lajos és Fornszek Sándor
jegyesei, kik habár hajadonok, hiven megvárták kedveseiket.
1854. áprilhó utolsó napjaiban értésünkre adták, hogy fogságunk
helyeül Józsefstadt tüzetett ki. Hová nem sokára el is inditottak
mindnyájunkat.
VI.
Fehérvártól Józsefstadtig.
fogságomban.