文档库 最新最全的文档下载
当前位置:文档库 › CHAPTER 1 Gene Trees, Species Trees, and Species Networks

CHAPTER 1 Gene Trees, Species Trees, and Species Networks

CHAPTER 1 Gene Trees, Species Trees, and Species Networks
CHAPTER 1 Gene Trees, Species Trees, and Species Networks

CHAPTER1

Gene Trees,Species Trees,and Species

Networks

Luay Nakhleh Derek Ruths

Department of Computer Science

Rice University

Houston,TX77005,USA

{nakhleh,druths}@https://www.wendangku.net/doc/0d14096912.html,

Hideki Innan

University of Texas Health Science Center at Houston

Houston,TX77030,USA

Hideki.Innan@https://www.wendangku.net/doc/0d14096912.html,

1.1Introduction

The availability of whole-genome sequence data has provided a rich resource of deep insights into many biological,medical and pharmaceutical problems and ap-plications,and is promising even more.Yet,along with these insights and promises, genomic data have given rise to many challenging problems,mainly due to the quan-tity and heterogeneity of such data.One of these major challenges is the phylogenetic analysis of multiple gene datasets that whole genomes provide.

Phylogeny,i.e.,the evolutionary history of a set of organisms,has become an indis-pensable tool in the post-genomic era.Emerging techniques for handling essential biological tasks(e.g.,gene?nding,comparative genomics,and haplotype inference) are usually guided by an underlying phylogeny.The performance of these techniques, therefore,depends heavily on the quality of the phylogeny.Almost all phylogenetic methods,however,assume that evolution is a process of strict divergence that can be modeled by a phylogenetic tree.While the tree model gives a satisfactory?rst-order approximation for many families of organisms,other families exhibit evolutionary events that cannot be represented by a tree.In particular,the evolutionary history of bacterial genomes is characterized by the occurrence of processes such as hor-izontal gene transfer(HGT)—transfer of genetic material across the boundaries of of distantly related species—and inter-speci?c recombination—exchange of genetic material.Further,hybrid speciation occurs among various groups of plants,?sh,and

1

2GENE TREES,SPECIES TREES,AND SPECIES NETWORKS frogs.In the presence of such evolutionary processes,the evolutionary relationship of a set of organisms is modeled by a phylogenetic network.

Accurate reconstruction of these processes bears signi?cant impact on many do-mains.The Tree of Life—the phylogeny of all organisms on Earth—is one of the grand challenges in evolutionary biology.The prokaryotic branch of this tree is be-lieved to have a large number of horizontal gene transfer events,in addition to re-combination events.Efforts to reconstruct a phylogeny for the prokaryotic branch may prove futile without developing phylogenetic network models and reconstruc-tion methods.

A signi?cant aspect of these complex evolutionary mechanisms is their contribu-tion to microbial genome diversi?cation.Like all forms of life,bacteria undergoes evolution.However,unlike many other organisms,bacterial evolution is not one of strict divergence.Recombination usually occurs within populations;in bacteria,how-ever,recombination occurs among different strains.Further,HGT is ubiquitous in the prokaryotic branch of the Tree of Life.Ho(2002)has recently written of the various health risks that recombination and HGT pose,including:(1)antibiotic resistance genes spreading to pathogenic bacteria,(2)disease-associated genes spreading and recombining to create new viruses and bacteria that cause diseases,and(3)transgenic DNA inserting into human cell,triggering cancer.Hence,detecting and reconstruct-ing these processes in bacteria play a major role in developing effective antibiotics, and bears a great impact on human health.

Biologists have long acknowledged the presence of these processes,their signi?-cance,and their effects.The computational research community has responded in recent years and proposed a plethora of methods for reconstructing complex evolu-tionary histories.The general theme of most existing methods can be summarized by:construct gene trees and reconcile them(this is known as the separate analysis approach).Gene tree reconciliation presents two major issues,namely identifying the(biological)source of incongruence,and(computationally)reconciling the trees. Many processes may lead to incongruent gene trees:

(1)Stochastic factors,such as wrong assumptions,insuf?cient data,incomplete sam-pling,and differential rates of sequence evolution across lineages.These factors do not violate the tree model of organismal evolutionary relationships;rather,the incon-gruence they cause must be eliminated in the early stages of phylogenetic analyses.

(2)Intra-species factors,such as gene loss and duplication.Although these events may lead to incongruent gene trees,they do not violate the tree model of organismal evolutionary relationships.

(3)Inter-species factors,such as horizontal gene transfer(whose rate is very high among prokaryotic organisms),and inter-speci?c recombination.These events result in networks of relationships,rather than trees of relationships.

In this work,we review the intra-and inter-species factors that cause gene tree incon-gruence and discuss current approaches for resolving these phenomena,with focus on non-treelike evolution.Further,we address extensions to the coalescent model to address non-treelike evolution.The rest of the chapter is organized as follows.In

GENE TREE INCONGRUENCE3 Section1.2we illustrate some of the processes that lead to incongruence gene trees. In Section1.3we review existing methods for addressing gene tree incongruence caused by gene loss and duplication(intra-species factors).In Section1.4,we de-scribe the phylogenetic network model and discuss the problem of reconciling gene trees into species networks.In Section1.5we propose approaches for extending the coalescent model to incorporate non-treelike evolutionary processes.We conclude the chapter in Section1.6.

1.2Gene Tree Incongruence

A gene tree is a model of how a gene evolves through duplication,loss,and nu-cleotide substitution.As a gene at a locus in the genome replicates and its copies are passed on to more than one offspring,branching points are generated in the gene tree.Because the gene has a single ancestral copy,barring recombination,the re-sulting history is a branching tree(Maddison(1997)).Sexual reproduction and mei-otic recombination within populations break up the genomic history into many small pieces,each of which has a strictly treelike pattern of descent(Hudson(1983b); Hein(1990);Maddison(1995)).Thus,within a species,many tangled gene trees can be found,one for each nonrecombined locus in the genome.A species tree de-picts the pattern of branching of species lineages via the process of speciation.When reproductive communities are split by speciation,the gene copies within these com-munities likewise are split into separate bundles of descent.Within each bundle,the gene trees continue branching and descending through time.Thus,the gene trees are contained within the branches of the species phylogeny(Maddison(1997)).

Gene trees can differ from one another as well as from the species tree.Disagree-ments(incongruence)among gene trees may be an artifact of the data and/or meth-ods used(stochastic factors).Various studies show the effects of stochastic factors on the performance of phylogenetic tree reconstruction methods(e.g.,Hillis et al. (1993);Hillis&Huelsenbeck(1994,1995);Nakhleh et al.(2001a,b,2002);Moret et al.(2002)).Stochastic factors confound the accurate reconstruction of evolution-ary relationships,and must be handled in the?rst stage of a phylogenetic analysis. Incongruence among gene trees due to intra-or inter-species processes,on the other hand,is a re?ection of true biological processes,and must be handled as such. Whereas eukaryotes evolve mainly though lineal descent and mutations,bacteria ob-tain a large proportion of their genetic diversity through the acquisition of sequences from distantly related organisms,via horizontal gene transfer(HGT)or recombi-nation(Ochman et al.(2000)).Views as to the extent of HGT and recombination in bacteria vary between the two extremes,with most views being in the middle (Doolittle(1999b,a);Kurland et al.(2003);et al.(2002);Hao&Golding(2004); et al.(2004);Nakamura et al.(2004)).However,there is a common belief that re-combination and HGT,among other processes,form the essence of prokaryotic evo-lution.Further,these two are the main processes(in addition to random mutations) by which bacteria develop resistance to antibiotics(e.g.,Lewis(1995);Ho(2002);

4GENE TREES,SPECIES TREES,AND SPECIES NETWORKS A B C g a g b g c A B C g a g b g c (a)(b)A B C g a g b g c Duplication

A B C g a g b g

c

(c)(d)

A B C g a g

b g

c G a G b G

c

(e)

Figure 1.1(a)Gene tree that agrees with the species tree.(b)Gene tree that disagrees with the species tree due to gene loss and duplication.(c)Gene tree that disagrees with the species tree due to HGT.(d)An inter-speci?c recombination event in which genetic material is exchanged between species B and C .(e)A hybrid speciation event that leads to two incongruent gene trees.

Enright et al.(2002);Paulsen et al.(2003)).Gene transfer and exchange are consid-ered a primary explanation of incongruence among bacterial gene phylogenies and a signi?cant obstacle to reconstructing the prokaryotic branch of the Tree of Life (Daubin et al.(2003)).

We illustrate some of the scenarios that may lead to gene tree incongruence in Figure

1.1.The species (or,organismal)tree is represented by the “tubes”;it has A and B as sister taxa whose most recent common ancestor (MRCA)is a sister taxon of C .Figure 1.1(a)shows a gene evolving within the branches of the same species tree;in this case,the topologies of the gene and species trees agree (the topology of this gene tree is shown in Figure 1.2(a)).In Figure 1.1(b)we show an example of how intra-species processes may lead to incongruent gene trees.The ?gure shows a gene evolving within the branches a species tree with one duplication event and three losses.Note that the species tree differs from the gene tree;based on this gene,B

GENE TREE INCONGRUENCE 5

g b g c g a G c

G b G a (a)(b)(c)(d)

Figure 1.2(a)The tree of the gene whose evolution is shown in Figure 1.1(a),and Figure

1.1(e).(b)The tree of the genes whose evolution is shown in Figures 1.1(b)and 1.1(c).(c)The tree of the gene involved in the recombination event shown in Figure 1.1(d).(e)The tree of the gene involved in the hybrid speciation event shown in Figure 1.1(e).

and C are sister taxa and their MRCA is a sister of taxon A .This gene tree is shown in Figure 1.2(b).

Another event that may cause incongruence between the species tree and the gene tree is HGT.In the case of HGT,shown in Figure 1.1(c),genetic material is trans-ferred from one lineage to another.Sites that are not involved in a horizontal transfer are inherited from the parent (as in Figure 1.2(a)),while other sites are horizontally transferred from another species (as in Figure 1.2(b)).

In the case of inter-speci?c recombination,as illustrated in Figure 1.1(d),some ge-netic material is exchanged between pairs of species;in this example,species B and C exchange genetic material.The genes involved in this exchange have an evolu-tionary history (shown in Figure 1.2(c))that is incongruent with that of the species.In hybrid speciation,two lineages recombine to create a new species.We can dis-tinguish diploid hybridization ,in which the new species inherits one of the two ho-mologs for each chromosome from each of its two parents—so that the new species has the same number of chromosomes as its parents,and polyploid hybridization ,in which the new species inherits the two homologs of each chromosome from both parents—so that the new species has the sum of the numbers of chromosomes of its parents.Under this last heading,we can further distinguish allopolyploidization ,in which two lineages hybridize to create a new species whose ploidy level is the sum of the ploidy levels of its two parents (the expected result),and auto-polyploidization ,a regular speciation event that does not involve hybridization,but which doubles the ploidy level of the newly created lineage.Prior to hybridization,each site on each homolog has evolved in a tree-like fashion,although,due to meiotic recombination,different strings of sites may have different histories.Thus,each site in the homologs of the parents of the hybrid evolved in a tree-like fashion on one of the trees induced by (contained inside)the network representing the hybridization event.Figure 1.1(e)shows a network with one hybrid.Each site evolves down exactly one of the two trees shown in Figures 1.2(a)and 1.2(d).

Notice that in the case of intra-species processes (gene loss and duplication),infer-ring the species tree from a set of potentially con?icting gene trees is a problem of reconciling the gene trees and explaining their differences through duplications and losses of genes.Therefore,in this case,despite the potential incongruence among the species and gene trees,the species phylogeny is still a tree (Mirkin et al.(1995);Page

6GENE TREES,SPECIES TREES,AND SPECIES NETWORKS (1990,1994);Eulenstein et al.(1998)).However,in the case of recombination,HGT, and hybrid speciation,the evolutionary history of the organismal genomes cannot be represented by phylogenetic trees;rather,phylogenetic networks are the appropriate model(Hallett&Lagergren(2001);Moret et al.(2004)).

1.3Trees Within Trees:The Gene Tree Species Tree Problem

Various reports of instances and effects of gene loss and duplication exist in the literature(e.g.,Moore(1995);Nichols(2001);Ruvolo(1997)).When losses and duplications are the only processes acting on the genes,a mathematical formulation of the gene tree reconciliation problem is as follows:

De?nition1.1(The Gene Tree Reconciliation Problem)

Input:Set T of rooted gene trees,a cost w D for duplications,and a cost w L for losses.

Output:Rooted tree T with each gene tree t∈T mapped onto T,so as to mini-mize the sum w D n D+w L n L,where n D is the total number of duplications and n L is the total number of losses,over all genes.

This problem was shown to be NP-hard by Fellows et al.(1998)and Ma et al. (1998).Heuristics for the problem exist,but these do not solve the optimization problem(see Ma et al.(1998);Page&Charleston(1997a)).Various?xed-parameter approaches have been proposed by Stege(1999a);Hallett&Lagergren(2000)and some variants can be approximated to within a factor of2and shown by Ma et al. (1998).

When loss and duplication are the only processes acting on the genes,two different questions can be posed,depending on the input data:

1.Gene tree reconciliation problem—when the gene trees are known and the species

tree is known,what is the best set of duplication and loss events that reconcile each gene tree with the species tree?

2.Species tree construction problem—when the gene trees are known,but the evolu-

tionary relationships among the species involved is not known,can the gene trees provide the information necessary to derive an estimate of the species tree? Both of these questions require the assumption of a certain model of gene duplication and loss.The complexity of the gene-tree reconciliation problem is determined by the model chosen,whereas the general species tree construction problem is NP-hard under all commonly used models of gene duplication and loss.

The simplest version of either problem uses a duplication-only model(i.e.,losses do not occur).During the period between years1995and2000,this was a commonly used model(Eulenstein et al.(1996);Page&Charleston(1997b);Page(1998);Eu-lenstein(1997);Stege(1999b);Ma et al.(1998);Zhang(1997);Ma et al.(2000)).

TREES WITHIN NETWORKS:THE GENE TREE SPECIES NETWORK PROBLEM7 Under the duplication-only model,the gene tree reconciliation problem has linear-time solutions(Zhang(1997);Eulenstein(1997)),as well as other polynomial-time algorithms that report better performance on real biological datasets(Zmasek&Eddy (2001)).The species tree construction problem is NP-hard,as was shown by Ma

et al.(1998).Different approaches have been taken to solving the species tree con-struction problem including heuristics(Page&Charleston(1997b)),approximation algorithms(Ma et al.(2000)),and?xed parameter tractable algorithms obtained by parameterizing by the number of gene duplications separating a gene tree from the species tree(Stege(1999b)).

The other common model used is the more general duplication-loss model,which admits both duplication and loss events within gene trees.The gene tree reconcil-iation problem has been shown to be polynomial-time under conditions where the evolution of the sequences themselves are not considered(Arvestad et al.(2004); Chen et al.(2000);Durand et al.(2005));if this is taken into account,the prob-lem becomes NP-hard(Fellows et al.(1998);Ma et al.(1998)).Various ef?cient heuristics for the problem are currently available(Arvestad et al.(2003,2004)). Early work on the gene tree reconciliation problem under this model borrowed tech-niques from biogeography and host/parasite evolution(Charleston(2000);Page& Charleston(1998)).

1.4Trees Within Networks:The Gene Tree Species Network Problem

As described in Section1.2,when events such as horizontal gene transfer,hybrid speciation,or recombination occur,the evolutionary history can no longer be mod-eled by a tree;rather,phylogenetic networks are the appropriate model in this case.

In this section,we describe the phylogenetic network model and approaches for re-constructing networks from gene trees.

1.4.1Terminology and notation

Given a(directed)graph G,let E(G)denote the set of(directed)edges of G and

V(G)denote the set of nodes of G.Let(u,v)denote a directed edge from node

u to node v;u is the tail and v the head of the edge and u is a parent of v.The indegree of a node v is the number of edges whose head is v,while the outdegree of

v is the number of edges whose tail is v.A node of indegree0is a leaf(often called

a tip by systematists).A directed path of length k from u to v in G is a sequence

u0u1···u k of nodes with u=u0,v=u k,and?i,1≤i≤k,(u i?1,u i)∈E(G);

we say that u is the tail of p and v is the head of p.Node v is reachable from u in G,denoted u;v,if there is a directed path in G from u to v;we then also say that u is an ancestor of v.A cycle in a graph is a directed path from a vertex back

to itself;trees never contain cycles:in a tree,there is always a unique path between two distinct vertices.Directed acyclic graphs(or DAGs)play an important role on

our model;note that every DAG contains at least one vertex of indegree0.A rooted

8GENE TREES,SPECIES TREES,AND SPECIES NETWORKS directed acyclic graph,in the context of this paper,is then a DAG with a single node of indegree0,the root;note that all all other nodes are reachable from the root by a(directed)path of graph edges.We denote by r(T)the root of tree T and by

L(T)the leaf set of T.Let T be a rooted phylogenetic tree over set S of taxa,and let S ?S.We denote by T(S )the minimal rooted subtree of T that connects all the element of S .Furthermore,the restriction of T to S ,denote T|S is the rooted subtree that is obtained from T(S )by suppressing all vertices(except for the root) whose number of incident edges is2.Let S be a maximum-cardinality set of leaves such that T1|S =T2|S ,for two trees T1and T2;we call T1|S (equivalently,T2|S ) the maximum agreement subtree of the two trees,denoted MAST(T1,T2).A clade of a tree T is a complete subtree of T.Let T =MAST(T1,T2);then,T1?T is the set of all maximal clades whose pruning from T1yields T (we de?ne T2?T similarly).In other words,there do not exist two clades u and u in T1?T such that either u is a clade in u ,or u is a clade in u.

We say that node x reaches node y in tree T if there is a directed path from x to y in T.We denote the root of a clade t by r(t).We say that clade t1reaches clade t2(both in tree T)if r(t1)reaches r(t2).The sibling of node x in tree T is node y,denoted sibling T(x)=y whenever x and y are children of the same node in T.We denote by T x the clade rooted at node x in T.The least common ancestor of a set X of taxa in tree T,denoted lca T(X)is the root of the minimal subtree of T that contains the leaves of X.The edge incoming into node x in tree T is denoted by inedge T(x).

1.4.2Phylogenetic networks

Moret et al.(2004)modeled phylogenetic networks using directed acyclic graphs (DAGs),and differentiated between“model”networks and“reconstructible”ones. Model networks A phylogenetic network N=(V,E)is a rooted DAG obeying certain constraints.We begin with a few de?nitions.

De?nition1.2A node v∈V is a tree node if and only if one of these three condi-tions holds:

?indegree(v)=0and outdegree(v)=2:root;

?indegree(v)=1and outdegree(v)=0:leaf;or

?indegree(v)=1and outdegree(v)=2:internal tree node.

A node v is a network node if and only if we have indegree(v)=2and outdegree(v)= 1.

Tree nodes correspond to regular speciation or extinction events,whereas network nodes correspond to reticulation events(such as hybrid speciation and horizontal gene transfer).We clearly have V T∩V N=?and can easily verify that we have V T∪V N=V.

TREES WITHIN NETWORKS:THE GENE TREE SPECIES NETWORK PROBLEM9 De?nition1.3An edge e=(u,v)∈E is a tree edge if and only if v is a tree node;

it is a network edge if and only if v is a network node.

The tree edges are directed from the root of the network towards the leaves and the network edges are directed from their tree-node endpoint towards their network-node endpoint.

A phylogenetic network N=(V,E)de?nes a partial order on the set V of nodes. We can also assign times to the nodes of N,associating time t(u)with node u;such

an assignment,however,must be consistent with the partial order.Call a directed path p from node u to node v that contains at least one tree edge a positive-time directed path.If there exists a positive-time directed path from u to v,then we must have t(u)

t(u)=t(v),because a reticulation event is effectively instantaneous at the scale of evolution;thus reticulation events act as synchronization points between lineages.

De?nition1.4Given a network N,two nodes u and v cannot co-exist(in time)if there exists a sequence P= p1,p2,...,p k of paths such that:

?p i is a positive-time directed path,for every1≤i≤k;

?u is the tail of p1,and v is the head of p k;and

?for every1≤i≤k?1,there exists a network node whose two parents are the head of p i and the tail of p i+1.

Obviously,if two nodes x and y cannot co-exist in time,then a reticulation event between them cannot occur.

De?nition1.5A model phylogenetic network is a rooted DAG obeying the follow-

ing constraints:

1.Every node has indegree and outdegree de?ned by one of the four combinations

(0,2),(1,0),(1,2),or(2,1)—corresponding to,respectively,root,leaves,inter-nal tree nodes,and network nodes.

2.If two nodes u and v cannot co-exist in time,then there does not exist a network

node w with edges(u,w)and(v,w).

3.Given any edge of the network,at least one of its endpoints must be a tree node.

Reconstructible networks De?nition1.5of model phylogenetic networks assumes that complete information about every step in the evolutionary history is available. Such is the case in simulations and in arti?cial phylogenies evolved in a laboratory setting—hence our use of the term model.When attempting to reconstruct a phylo-genetic network from sample data,however,a researcher will normally have only incomplete information,due to a combination of extinctions,incomplete sampling, and abnormal model conditions.Extinctions and incomplete sampling have the same consequences:the data do not re?ect all of the various lineages that contributed to

the current situation.Abnormal conditions include insuf?cient differentiation along

10GENE TREES,SPECIES TREES,AND SPECIES NETWORKS edges,in which case some of the edges may not be reconstructible,leading to poly-tomies and thus to nodes of outdegree larger than2.All three types of problems may lead to the reconstruction of networks that violate the constraints of De?nition1.5. (The distinction between a model phylogeny and a reconstructible phylogeny is com-mon with trees as well:for instance,model trees are always rooted,whereas recon-structed trees are usually unrooted.In networks,both the model network and the reconstructed network must be rooted:reticulations only make sense with directed edges.)Clearly,then,a reconstructible network will require changes from the de?ni-tion of a model network.In particular,the degree constraints must be relaxed to allow arbitrary outdegrees for both network nodes and internal tree nodes.In addition,the time coexistence property must be reconsidered.

There are at least two types of problems in reconstructing phylogenetic networks. First,slow evolution may give rise to edges so short that they cannot be reconstructed, leading to polytomies.This problem cannot be resolved within the DAG framework,

so we must relax the constraints on the outdegree of tree nodes.Secondly,missing data may lead methods to reconstruct networks that violate indegree constraints or time coexistence.In such cases,we can postprocess the reconstructed network to restore compliance with most of the constraints in the following three steps:

1.For each network node w with outdegree larger than1,say with edges(w,v1),

...,(w,v k),add a new tree node u with edge(w,u)and,for each i,1≤i≤k, replace edge(w,v i)by edge(u,v i).

2.For each network node w whose parents u and v violate time coexistence,add

two tree nodes w u and w v and replace the two network edges(u,w)and(v,w) by four edges:the two tree edges(u,w u)and(v,w v)and the two network edges (w u,w)and(w v,w).

3.For each edge(u,v)where both u and v are network nodes,add a new tree node

w and replace the edge(u,v)by the two edges(u,w)and(w,v).

The resulting network is consistent with the original reconstruction,but now satis?es

the outdegree requirement for network nodes,obeys time coexistence(the introduc-tion of tree edges on the paths to the network node allow arbitrary time delays),and

no longer violates the requirement that at least one endpoint of each edge be a tree node.Moreover,this postprocessing is unique and quite simple.We can thus de?ne

a reconstructible network in terms similar to a model network.

De?nition1.6A reconstructible phylogenetic network is a rooted DAG obeying the following constraints:

1.Every node has indegree and outdegree de?ned by one of the three(indegree,outdegree)

combinations(0,x),(1,y),or(z,1),for x≥1,y≥0,and z≥2—corresponding to,respectively,root,other tree nodes(internal nodes and leaves),and network nodes.

2.If two nodes u and v cannot co-exist in time,then there does not exist a network

node w with edges(u,w)and(v,w).

3.Given any edge of the network,at least one of its endpoints must be a tree node.

TREES WITHIN NETWORKS:THE GENE TREE SPECIES NETWORK PROBLEM11 De?nition1.7A network N induces a tree T if T can be obtained from N by the following two steps:

1.For each network node in N,remove all but one of the edges incident into it;and

2.for every node v such that indegree(v)=outdegree(v)=1,the parent of v

is u,and the child of v is w,remove v and the two edges(u,v)and(v,w),and

add new edge(u,w)(this is referred to in the literature as the forced contraction

operation).

For example,the network N shown in Figure1.1(e)induces both trees shown in Figure1.2(a)and Figure1.2(d).

1.4.3Reconstructing networks from gene trees

From a graph-theoretic point of view,the problem can be formulated as pure phylo-genetic network reconstruction(Moret et al.(2004);Nakhleh et al.(2004,2005)).

In the case of HGT,and despite the fact the evolutionary history of the set of organ-isms is a network,Lerat et al.(2003)showed that an underlying species tree can

still be inferred.In this case,a phylogenetic network is a pair(T,Ξ),where T is the species(organismal)tree,andΞis a set of HGT edges whose addition to T results

in a phylogenetic network N that induces all the gene trees.The problem can be formulated as follows.

De?nition1.8(The HGT Reconstruction Problem)

Input:A species tree ST and a set G of gene trees.

Output:SetΞof minimum cardinality whose addition to ST yields phylogenetic

network N that induces each of the gene trees in G.

However,in the case of hybrid speciation,there is no underlying species tree;instead

the problem is one of reconstructing a phylogenetic network N that induces a given

set of gene trees.

De?nition1.9(The Hybrid Speciation Reconstruction Problem)

Input:A set G of gene trees.

Output:A Phylogenetic network N with minimum number of network nodes that

induces each of the gene trees in G.

The minimization criterion re?ects the fact that the simplest solution is sought;in

this case,the simplest solution is one with the minimum number of HGT or hybrid speciation events.We illustrate this point with the example species tree ST in Figure

1.3(a)and the gene tree GT in Figure1.3(b).Assume that the actual HGT events that

took place are the one depicted in Figure1.3(c).Nonetheless,the scenario depicted

12GENE TREES,SPECIES TREES,AND SPECIES NETWORKS

(a)(b)(c)(d)

Figure1.3(a)A species tree ST.(b)A tree GT of a horizontally transferred gene.Both net-works in(c)and(d)are formed based on ST,and both induce GT.However,even though the actual HGT scenarios that took place are described by the network in(c),the HGT Recon-struction Problem seeks the solution in(d).

in Figure1.3(d)has fewer HGT events,yet induces GT.In this case,the solution in Figure1.3(d)is the one sought by the HGT Reconstruction Problem.Although the scenarios depicted in Figure1.3(c)and Figure1.3(d)are very different,inferring the one in Fig1.3(c)as the correct solution,in the absence of any additional biologi-cal knowledge about the organisms,would be rather arbitrary.Hence,based on the species and gene tree topologies,solving the HGT Reconstruction Problem offers the “best”solution.Another serious problem that impacts the identi?ability of reticulate evolution is that of extinction and incomplete taxon sampling.Moret et al.(2004) illustrated some of the scenarios that lead to non-identi?ability of reticulation events from a set of gene trees.

Hallett&Lagergren(2001)gave an ef?cient algorithm for solving the HGT Re-construction Problem;however,their algorithm handles limited special cases of the problem in which the number of HGT events is very small,and the number of times a gene is transferred is very low(also,their tool handles only binary trees;Addario-Berry et al.(2003)).Nakhleh et al.(2004)gave ef?cient algorithms for solving the Hybrid Speciation Reconstruction Problem,but for constrained phylogenetic net-works,referred to as gt-networks;further,they handled only binary trees.Nakhleh et al.(2005)have recently introduced RIATA-HGT,which is the?rst method for solving the general case of the HGT Reconstruction Problem.The method can be easily modi?ed to yield a heuristic for solving the Hybrid Speciation Reconstruction Problem.We now describe the method and its empirical performance.

RIATA-HGT:reconstructing HGT from gene trees

We describe the algorithm underlying RIATA-HGT in terms of a species tree and a gene tree.The core of RIATA-HGT is the divide-and-conquer algorithm Com-puteHGT algorithm(outlined in Figure1.4).The algorithm starts by computing the MAST,T ,of the species tree ST and gene tree GT;tree T forms the basis for de-tecting and reconstructing the HGT events(computing T is done in Step1in Figure

1.4).The algorithm then decomposes the clade sets U1and U2(whose removal from

TREES WITHIN NETWORKS:THE GENE TREE SPECIES NETWORK PROBLEM13 ST and GT,respectively,yields T )into maximal clades such that each maximal clade in one of the two sets is“matched”by a maximal clade on the same leaf set in

the second set.The algorithm for this decomposition is outlined in Figure1.5.The al-gorithm then recurses on each maximal clade and its matching maximal clade(Steps

5.c.(1)and5.d.(5).(1)in Figure1.4)to compute the HGT events whose recipients form sub-clades of those maximal clades.Finally,we add a single HGT event per each maximal clade to connect it to its“donor”in the ST;this is achieved through

the calls to AddSingleHGT(Figure1.6)in Steps5.c.(2)and5.d.(5).(3)in Figure1.4.

P ROCEDURE C OMPUTE HGT(ST,GT)

Input:Species tree ST,and gene tree GT,both on the same set S of taxa.

Output:Computes the setΞof HGT events such that the pair(ST,Ξ)induces

GT.

1.T =MAST(ST,GT);

2.If T =ST then

(a)Return;

3.U1=ST?T ;U2=GT?T ;

4.V=?;

5.Foreach u2∈U2

(a)Decompose(U1,u2,T ,V);

6.U2=V;

7.While V=?

(a)Let u2be an element of V;

(b)Let u1∈U1be such that L(u2)?L(u1);

(c)Y={y∈U2:L(y)∩L(u1)=?};

(d)Z={y|(L(y)?L(u1)):y∈Y};

(e)V=V?Y;V=V∪Z;

(f)X={u1|L(y):y∈Y};

(g)Foreach y∈Y

i.Let x∈X be such that L(x)∩L(y)=?;

https://www.wendangku.net/doc/0d14096912.html,puteHGT(x,y);

iii.AddSingleHGT(ST,GT,y,U2,T );

Figure1.4The main algorithm for detecting and reconstructing HGT events based on a pair

of species tree and gene tree.

Theoretically,RIATA-HGT may not compute the minimum-cardinality set of HGT events;Nakhleh et al.(2005)established the following properties of the method. Theorem1.1Given a species tree ST and a gene tree GT,the network N ob-

14GENE TREES,SPECIES TREES,AND SPECIES NETWORKS P ROCEDURE D ECOMPOSE(U1,u2,T,U )

Input:Set U1of clades from ST,clade u2from GT,the backbone clade u2,and U which will contain the“re?ned”clades of u2.

Output:Decompose u2so that no clade in U has a leaf set that is the union of leaf sets of more than one clade in U1.

1.If?u1∈U1such that L(u2)?L(u1)then

(a)U =U ∪{u2};

(b)B(u2)=T;

(c)Return u2;

2.Else

(a)If?u1∈U1such that r(u2)=r(u2|L(u1))

i.t=u2|L(u1);

ii.U =U ∪{t};

iii.B(t)=T;

iv.Let X=u2?t;

v.Foreach x∈X

A.Decompose(U1,x,t,U );

vi.Return t;

(b)Else

i.Let c1,...,c k be the children of r(u2);

ii.x=Decompose(U1,T c

1,T,U );

iii.For i=2to k

A.Decompose(U1,T c

i

,x,U );

iv.Return x;

Figure1.5The algorithm for decomposing the clades in U1and U2such that for all u1∈U1 and u2∈U2we have L(u1)?L(u2).

tained by running RIATA-HGT on(ST,GT)induces GT.Further,RIATA-HGT takes O(n4)time in the worst case,where n is the number of leaves in ST.

Moreover,experimental results show very good empirical performance on synthetic data,as illustrated in Figure1.7.The whisker-and-box plot in Figure1.7(a)shows the individual numbers of HGT events as predicted by RIATA-HGT versus the ac-tual numbers.Figure1.7(b)shows the average(of30runs)numbers of HGT events as predicted by RIATA-HGT versus the actual numbers(for full details of how the simulation studies were conducted and detailed analyses of the results,please re-fer to Nakhleh et al.(2005)).The plots demonstrate empirically the excellent per-formance of RIATA-HGT;it estimates the exact number of HGT events in a great majority of the cases,with very mild over-or under-estimation in the other cases. Over-estimation is an artifact of the heuristic nature of RIATA-HGT,whereas under-estimation is an artifact of the parsimony criterion in the de?nition of the prob-

TREES WITHIN NETWORKS:THE GENE TREE SPECIES NETWORK PROBLEM 15P ROCEDURE A DD S INGLE HGT(ST ,GT ,u 2,U 2,T )

Input:Species tree ST ,gene tree GT ,clade u 2of GT ,set U 2of clades of GT ,and MAST T of ST and GT .

Output:Add to Ξa single HGT event whose donor is determined in this procedure and whose recipient is clade u 2.

1.Q =L (u 2)∪L (B (u 2));

2.T =GT |Q ;p =lca T

(L (u 2));

3.tq =lca ST (L (u 2));te

=inedge ST (tq );

4.If p is a child of r (T )and |L (B (u 2))|>1then

(a)sq =lca ST (L (B (u 2)));

(b)Ξ=Ξ∪(sq →te );

5.Else

(a)O =

{p :p =sibling T (p )}L (T p );

(b)sq =lca ST (O );se =inedge ST (

sq );

(c)Ξ=Ξ∪(se →te );

Figure 1.6The algorithm for detecting and reconstructing the single HGT event in which clade u 2is the recipient.

lem (see the discussion above).RIATA-HGT was also applied to the bacterial gene

(a)(b)

Figure 1.7The results of RIATA-HGT on synthetic datasets.(a)A box-and-whisker plot for the predictions of HGT event numbers made by RIATA-HGT.(b)The averages of HGT event numbers estimated by RIATA-HGT vs.the actual number of HGT events.Each point is the average of 30runs of RIATA-HGT.

datasets reported in Lerat et al.(2003),and produced the results hypothesized by Lerat et al.In summary,RIATA-HGT performed very well on the synthetic datasets,as well as on the biological datasets.

16GENE TREES,SPECIES TREES,AND SPECIES NETWORKS 1.5The Coalescent and Reticulate Evolution

1.5.1The coalescent and lineage sorting in ancestral populations

Intra-species events(i.e.,gene duplication and loss)occur because of random contri-bution of each individual to the next generation.Some fail to have offsprings(gene loss)while some happen to have multiple offsprings(gene duplication).This means a number of duplication and loss events occur every generation.In population ge-netics,this process was?rst modeled by R.A.Fisher and S.Wright,in which each gene of the population at a particular generation is chosen independently from the gene pool of the previous generation,regardless of whether the genes are in the same individual or in different individuals.

Under the Wright-Fisher model,“the coalescent”considers the process backward in time(Kingman(1982);Hudson(1983b);Tajima(1983)).That is,the ancestral lineages of genes of interest are traced from offsprings to parents.A coalescent event occurs when two(or sometimes more)genes are originated from the same parent, which is called the most recent common ancestor(MRCA)of the two genes.This event corresponds to gene duplication when the process is considered forward in time.Gene loss events can be ignored in the coalescent process,because we are not interested in the lineages that do not exist at present.

The basic process can be treated as follows.Consider a pair of genes at timeτ1in a random mating haploid population.The population size at timeτis denoted by N(τ).The probability that the pair are from the same parental gene at the previous generation(timeτ1+1)is1/N(τ1+1).Therefore,starting atτ1,the probability that the coalescence between the pair occurs atτ2is given by

P rob(τ2)=

1

N(τ2)

τ2?1

τ=τ1+1 1N(τ) .(1.1)

When N(τ)is constant,the probability density distribution(pdf)of the coalescent time(i.e.,t=τ2?τ1)is given by a geometric distribution,and can be approximated by an exponential distribution for a large N:

P rob(t)=1

N

e?t/N.(1.2)

The coalescent process is usually ignored in phylogenetic analysis,but has a sig-ni?cant effect(causing lineage sorting)when closely related species are considered (Hudson(1983a);Takahata(1989);Rosenberg(2002)).The situation of Figure1.1(b) is reconsidered under the framework of the coalescent in Figure1.8.Here,it is as-sumed that species A and B split T1=5generations ago,and the ancestral species of A and B and species C split T2=19generation ago.The ancestral lineage of

a gene from species A and that from B meet in their ancestral population at time τ=6,and they coalesce atτ=35,which predates T2,the speciation time between (A,B)and C.The ancestral lineage of B enters in the ancestral population of the three species at timeτ=20,and?rst coalesces with the lineage of C.Therefore,

THE COALESCENT AND RETICULATE EVOLUTION 17

T

T T Figure 1.8An illustration of the coalescent process in a three species model with discrete generations.The process is considered backward in time from present,T 0,to past.Circles represent haploid individuals.We are interested in the gene tree of the three genes (haploids)from the three species.Their ancestral lineages are represented by closed circles connected by lines.A coalescent event occurs when a pair of lineages happen to share a single parental gene (haploid).

the gene tree is represented by A (BC )while the species tree is (AB )C .That is,the gene tree and species tree are “incongruent”.Under the model in Figure 1.8,the probability that the gene tree is congruent with the species tree is 0.85,which is one minus the product of the probability that the ancestral lineages of A and B do not coalesce between τ=6and τ=9,and the probability that the ?rst coalescence in the ancestral population of the three species occur between (A and C )or (B and C ).The former probability is 141512131112...7878=0.22and the latter is 23.

Under the three-species model (Figure 1.8),there are three possible types of gene tree,(AB )C ,(AC )B and A (BC ).Let P rob [(AB)C ],P rob [(AC)B ]and P rob [A(BC)]be the probabilities of the three types of gene tree.These three probabilities are sim-ply expressed with a continuous time approximation when all populations have equal and constant population sizes,N ,where N is large:

P rob [(AB)C ]=1?2

3e ?(T 2?T 1)/N ,(1.3)

and

P rob [(AC)B ]=P rob [A(BC)]=1

3e ?(T 2?T 1)/N .(1.4)

Figure 1.9(a)shows the three probabilities as functions of (T 2?T 1)/N .

An interesting application of this three species problem is in hominoids;A :human,B :chimpanzee and C :gorilla.It is believed that the species three is (AB )C .Chen &Li (2001)investigated DNA sequences from 88autosomal intergenic regions,and the gene tree is estimated for each region.They found that 36regions support the species tree,(AB )C ,while 10estimated trees are (AC )B and 6are A (BC ).No resolution is obtained for the remaining 36regions (see below).It is possible to estimate the time

18GENE TREES,SPECIES TREES,AND SPECIES NETWORKS

(a)(b)

Figure1.9(a)The probabilities of the three types of gene tree,(AB)C,(AC)B,and A(BC), as functions of(T2?T1)/N.(b)The probabilities that the gene tree is resolved from DNA sequence data.The probabilities are given functions of the mutation rate for the three types of tree,(AB)C,(AC)B,and A(BC),when(T2?T1)/N=0.5.The white regions represent the probabilities that the gene tree is not resolved.

between two speciation events,T2?T1,assuming all populations have equal and constant diploid population sizes,N(Wu(1991)).Since36out of52gene trees are congruent with the species tree,T2?T1is estimated to be?ln[(3/2)(36/52)]=0.77 times2N generations.It should be noted that2N is used for the coalescent time scale instead of N because hominoids are diploids.If we assume N to be5×104?1×105 (Takahata et al.(1995);Takahata&Satta(1997)),the time between two speciation events is7.7?15.5×104generations,which is roughly1?3million years assuming a generation time of15?20years.

It is important to notice that the estimation of the gene tree from DNA sequence data is based on the nucleotide differences between sequences,and that the gene tree is sometimes unresolved.One of the reasons for that is a lack of nucleotide differences such that DNA sequence data are not informative enough to resolve the gene tree. This possibility strongly depends on the mutation rate.Letμbe the mutation rate per region per generation,and consider the effect of mutation on the estimation of the gene tree.We consider the simplest model of mutations on DNA sequences,the in?nite site model(Kimura(1969)),in which mutation rate per site is so small that no multiple mutations at a single site are allowed.Consider a gene tree,(AB)C, and suppose that we have a reasonable outgroup sequence such that we know the sequence of the MRCA of the three sequences.It is obvious that mutations on the internal branch between the MRCA of the three and the MRCA of A and B are informative.If at least one mutation occurred on this branch,the gene tree can be resolved from the DNA sequence alignment.This effect is investigated by assuming that the number of mutations on a branch with length t follows a Poisson distribution with meanμt.Figure1.9(b)shows the probability that the gene tree is resolved; T2?T1=0.5N generations is assumed so that the probability that the gene tree is (AB)C is about0.6.As expected,as the mutation rate increases,the probability that the gene tree is resolved from the sequence alignment increases,and this probability

THE COALESCENT AND RETICULATE EVOLUTION19

(a)(b)

Figure1.10(a)A three species model with a HGT event.A demonstration that a congruent tree could be observed even with HGT.(b)The probabilities of the three types of gene tree, (ab)c’,(ac’)b,and a(bc’),as functions of T h/N.T1=2N and T2=3N are assumed.

exceeds90%when Nμ>1.52.Similar results are obtained for the other two types of trees,(AC)B and A(BC),that appears with probability0.2for each(see also Figure1.9(b)).

1.5.2Gene trees,species trees and reticulate evolution

In the previous section,we have shown that the gene tree is not always identical to the species tree.With keeping this in mind,let us consider the effect of horizontal gene transfer(HGT)on gene tree under the framework of the coalescent.

The application of the coalescent theory to bacteria is straightforward.Bacterial evo-lution is better described by the Moran model rather than the Wright-Fisher model because bacteria do not?t a discrete generation model.Suppose that each haploid in-dividual in a bacterial population with size N has a lifespan that follows an exponen-tial distribution with mean l.When an individual dies,another individual randomly chosen from the population replaces it to keep the population size constant.In other words,one of the N?1alive lineages is duplicated to replace the dead one.Un-der the Moran model,the ancestral lineages of individuals of interest can be traced backward in time,and the coalescent time between a pair of individuals follows an exponential distribution with mean lN/2(Ewens(1979);Rosenberg(2005)).This means that one half of the mean lifetime in the Moran model corresponds to one generation in the Wright-Fisher model.

It may usually be thought that HGT can be detected when the gene tree and species tree are incongruent(see Section1.4).However,the situation is complicated when lineage sorting is also involved.Consider a model with three species,A,B,and C, in which an HGT event occurs from species B to C.Suppose the ancient circular

20GENE TREES,SPECIES TREES,AND SPECIES NETWORKS genome has a single copy of a gene as illustrated in Figure 1.10(a).Let a ,b and c be the focal orthologous genes in the three species,respectively.At time T h ,a gene escaped from species B and was inserted in a genome in species C at T i ,which is denoted by c .Following the HGT event,c was physically deleted from the genome,so that each of the three species currently has a single copy of the focal gene.If there is no lineage sorting,the gene tree should be a (bc ).Since this tree is incon-gruent with the species tree,(AB )C ,we could consider it as an evidence for HGT.However,as demonstrated in Section 1.2,lineage sorting could also produce the in-congruence between the gene tree and species tree without HGT.It is also important to note that lineage sorting,coupled with HGT,could produce congruent gene tree,as illustrated in Figure 1.10(a).Although b and c have more chance to coalesce ?rst,the probability that the ?rst coalescence occurs between a and b or between a and c may not be negligible especially when T 1?T h is short.

The probabilities of the three types of gene tree can be formulated under this tri-species model with HGT as illustrated in Figure 1.10(a).Here,T h could exceed T 1,in such a case it can be considered that HGT occurred before the speciation between A and B .Assuming that all populations have equal and constant population sizes,N ,the three probability can be obtained

modifying (1.3)and (1.4):P rob [(AB)C ]= 1

3e ?(T 1?T h )/N if T

h ≤T 1

1?23e ?(T h ?T 1)/N if T h >T 1,(1.5)

P rob [(AC)B ]= 13e

?(T 1

?T h )/N if T

h

≤T 113e ?(T h ?T 1)/N if T h >T 1,(1.6)

and

P rob [A(BC)]= 1?2

3e ?(T 1?T h )/N if T h ≤T 113e ?(T h ?T 1)/N if T h >T 1.(1.7)

Figure 1.10(b)shows the three probability assuming T 1=2N and T 2=3N .

Thus,lineage sorting due to the coalescent process works as a noise for detecting and reconstructing HGT based on gene tree,sometimes mimicking the evidence for HGT and sometimes creating a false positive evidence for HGT.Therefore,to distinguish HGT and lineage sorting,statistics based on the theory introduced in this chapter is needed.We only considered very simple cases with three species here,but it is straightforward to extend the theory to more complicated models.

1.6Summary

In this chapter,we have reconsidered the gene tree species tree problem in the con-text of reticulate evolution.In particular,we discussed gene tree incongruence due to reticulate evolution and presented our recent heuristic,RIATA-HGT,for resolv-ing this type of incongruence.Further,we have addressed extensions of the coales-cent model to incorporate non-treelike evolutionary events,such as horizontal gene transfer.Gene tree incongruence is both an obstacle impeding accurate phylogeny

Chapter1会计概论答案

第一章会计透视:会计信息及其使用者▓复习思考题 1.概述会计的性质。 会计是按照会计规范确认、计量、记录一个组织的经济活动,运用特定程序处理加工经济信息,并将处理结果传递给会计信息使用者的信息系统,是组织和总结经济活动信息的主要工具。 会计是一个信息系统,会计处理的各个环节的加工的对象是会计信息。会计信息实际上是一种广义的信息,包含三个层次: 其一,以货币化指标体现的财务信息,它是从动态、静态两个角度,对特定主体经济资源的数量(资产)、归属(负债、所有者权益)、运用效果(收益分配)、增减变化及其结果(财务状况变动及其结果)进行描述; 其二,非货币化的和非数量化的说明性信息,它们不仅仅是对主体的财务状况、经营成果等财务指标的基本说明,而且还包含了大量的主体所处的社会、文化、道德、法律等环境信息,这些信息对于使用者正确判断主体的经营能力、发展前景,往往起到至关重要的作用; 其三,其他用于主体内部管理的信息,这些信息常常由成本会计、管理会计以及内部审计人员提供,主要包括了短期(长期)决策信息、预算信息、责任中心要求及履行情况等情况,虽然与外部性较强的财务会计信息相比,它们更容易为人们所忽略,但在经济管理和财务信息质量控制方面,它们也起了不可低估的作用。不过,在当前的会计报告模式中,所反映的会计信息主要是前两个层次上的信息。 2.企业的获利能力是否为债权人的主要考虑因素? 债权人关心那些影响自己的债权能否得到按期偿还的因素。他们会对公司的获利能力及清偿能力感兴趣。债权人会从获利能力去衡量未来的现金流量,由于企业的获利能力与现金流量并不一定同步产生,所以对短期债权人来说,企业的获利能力不能成为债权人的主要考虑因素,但对相对长期的债权人来说获利能力应该是值得更加关注的因素。 3.财务报表中体现出来的会计信息用以满足不同使用群体的需求,但并不是所有的使用者都能得到相同的满足的。在实践中,外部财务报表使用者诸如股东、供应商、银行等是如何获得有关公司的财务信息的?若要同时满足不同类别的财务报表使用者的信息需求有什么困难? 首先,股东是公司法定的所有者,法律上财务报表是为其制作的。法律规定,公司必须定期编制并发布财务报表,为其股东公布财务信息,我国《公司法》对此也有明确的规定。 其次,供应商与公司是商业伙伴关系,不提供商业信用的供应商,由于是钱货两讫,可通过报媒和网络了解公司公开财务的财务信息,如果供应商对公司提供商业信用,供应商可要求企业提供一定的财务信息,但对这些财务信息的准确性、可靠性需要有一定的分析。

(完整版)中国翻译史及重要翻译家

中国翻译史及重要翻译家 08英本1 杨慧颖 NO.35 中国翻译史上有许多为人们所熟知的大家,现就其翻译观点和主要作品做一简介: 严复 是中国近代翻译史上学贯中西、划时代意义的翻译家,也是我国首创完整翻译标准的先驱者。严复吸收了中国古代佛经翻译思想的精髓,并结合自己的翻译实践经验,在《天演论》译例言里鲜明地提出了“信、达、雅”的翻译原则和标准。“信”(faithfulness)是指忠实准确地传达原文的内容;“达”(expressiveness)指译文通顺流畅;“雅”(elegance)可解为译文有文才,文字典雅。这条[1]著名的“三字经”对后世的翻译理论和实践的影响很大,20世纪的中国译者几乎没有不受这三个字影响的。 主要翻译作品: 《救亡决论》,《直报》,1895年 《天演论》,赫胥黎,1896年~1898年 《原富》(即《国富论》),亚当·斯密,1901年 《群学肄言》,斯宾塞,1903年 《群己权界论》,约翰·穆勒,1903年 《穆勒名学》,约翰·穆勒,1903年 《社会通诠》,甄克斯,1903年 《法意》(即《论法的精神》),孟德斯鸠,1904年~1909年 《名学浅说》,耶方斯,1909年 鲁迅 鲁迅翻译观的变化,从早期跟随晚清风尚以意译为主,到后期追求直译、反对归化。鲁迅的翻译思想主要是围绕"信"和"顺"问题展开的。他"宁信而不顺"的硬译观在我国文坛上曾经引发过极大的争议。 鲁迅先生说过:凡是翻译,必须兼顾两面,一则当然力求其易解,一则是保存着原作的丰姿。从实质上来讲,就是要使原文的内容、风格、笔锋、韵味在译文中得以再现。翻译涉及原语(source language)与译语(target language) 两种语言及其文化背景等各方面的知识,有时非常复杂。所以,译者要想收到理想的翻译效果,常常需要字斟句酌,反复推敲,仅仅懂得一些基本技巧知识是不够的,必须广泛涉猎不同文化间的 差异,必须在两种语言上下工夫,乃至独具匠心。

中国翻译简史

百科知识—中国翻译简史 中国历史上曾经出现了四次翻译高潮: 1东汉至唐宋的佛经翻译 2明末清初的科技翻译 3“五四”时期的西学翻译 4中国历史上第四次翻译高潮(五四以后---当代) 古代佛经翻译(四个阶段) 第一阶段从东汉末年到西晋(起步或草创阶段) 主要是外籍僧人和华籍胡裔僧人,翻译主要靠直译,甚至是“死译”、“硬译”,采取 口授形式,因此可信度不高。 代表人物支谦 支谦,三国时佛经翻译家,又名支越,字恭明。支谦的译述比较丰富:约三十年间,译出佛经《大明度无极经》、《大阿弥陀经》等八十八部、一百一十八卷,创作了《赞菩萨连句梵呗》三契,其翻译以大乘“般若性空”为重点,为安世高、支谶以后译经大师。支谦自译的经也偶尔加以自注,像《大明度无极经》首卷,就是一例。这种 作法足以济翻译之穷,而使原本的意义洞然明白。 特点 1主要力量:外籍僧人和华籍胡裔僧人 2翻译全凭口授 3大多才有直译法 5佛经内容经常采用中国本土道家思想 古代佛经翻译 第二阶段从东晋到隋末(发展期) 释道安总结出了“五失本”、“三不易”的规律; 释道安,南北朝时高僧,翻译家。本姓卫,常山抚柳(河北冀州)人。总结 了汉代以来流行的佛教学说,整理了新译旧译的经典,编纂目录,确立戒规,主 张僧侣以“释”(释迦牟尼)为姓。主要监督翻译了《四阿含》等。主张直译,

不增不减,只做词序调整。翻译佛经在五种情况下会失去本来面目,有三件事决定了译事是很不容易的,因此必须慎之又慎。 彦琮在其论着《辩正论》中提出了翻译要例“十条和对译者的要求“八备”” 彦琮(557—610年),俗姓李,邢台隆尧县双碑人,隋代著名高僧,他精通梵文, 也是我国佛教史上屈指可数的佛经翻译家和佛教著作家。《辩正论》里翻译要例八备十条:就是对做翻译人的具体要求和翻译要求。 鸠摩罗什开始提倡意译; 鸠摩罗什(梵语Kumārajīva )(公元344 ~ 413 年),音译为鸠摩罗耆婆,又作鸠 摩罗什婆,简称罗什。其父名鸠摩罗炎,母名耆婆,属父母名字的合称,汉语的意思为“童寿”。东晋时后秦高僧,著名的佛经翻译家。与真谛(499—569)、玄奘(602~664)、并称为中国佛教三大翻译家。 慧远等人则对译文的风格和文体问题进行了一定的探索。 慧远,俗姓贾,雁门楼烦(约在今山西朔城区)人。出生于代州(约代县) 主张直译与意译、音译的结合,著有《三法度序》 真谛(449-569)梁陈时代人,翻译《摄大乘论》 古代佛经翻译(唐朝——全盛期) 古代佛经翻译(续) 主要译者多为本国人,他们除了精通佛理以外还精通梵汉两种语言,其译作在质量和数量上都大大超过了前两个阶段。加上唐朝统治者的重视和支持,这一阶段的译经活动达到顶峰,出现了玄奘、不空、义净等著名的僧人译经家。 玄奘提出了佛经翻译中著名的“五不翻”的原则,并为译经者进行了十一种详细的分工。 鸦片战争至甲午战争前 著名的禁烟英雄林则徐决心“师夷之长技以制夷”,开始“日日使人刺探西事,翻译西书”,被誉为“组织翻译活动的先驱”。语言学家马建忠在其《拟设翻译书院议》一文中留下了一些有关翻译的论述。他提出有三类书籍急需翻译,并指出“需要择其善者译之“。此外,他在总结了当时译文中常见的不足之后提出了所谓“善译”的翻译标准,要求译文与原文之间“无毫发出入于其间”。 近代翻译

中国翻译史1

History of Translation Teaching Plan Teaching Contents: 1. An introduction to Chinese translation theory and history 2. An introduction to western countries translation theory and history 中国翻译史的大致分期 1.由汉代到唐宋的上千年的佛经翻译【支谦、道安、鸠摩罗什、昙无谶、法显、谢灵运、真谛、彦琮、慧远、玄奘、不空】 2.明清交替之际的科技翻译【徐光启、利玛窦、汤若望、南怀仁、熊三拔、李之藻等】 3.清末民初的文学和科技翻译【李善兰、华蘅芳、傅兰雅、林纾、严复、梁启超等】 4. 民国时期的翻译【赵元任、朱生豪、林语堂】 5. 新中国成立后的翻译,尤其是改革开放以来的翻译【傅雷、钱钟书、杨绛】 Lecture 1 佛经翻译 I.关于翻译的早期记载 《册府元龟·外臣部·鞮(di)译》记载,周时有越裳国“以三相胥重译而献白雉,曰:‘道路悠远,山川阻深,音使不通,故重译而朝’”。 “五方之民,言语不通,嗜欲不同。达其志,通其欲,东方曰寄,南方曰象,西方曰狄鞮,北方曰译。”《礼记·王制》 翻译的不同称呼:“象寄”、“象胥”、“鞮译”“舌人” 寄send; entrust; rely on 象be like; resemble; image 译translate; interpret 越人歌 今夕何夕兮? 搴舟中流。 今日何日兮? 得与王子同舟。 蒙羞被好兮, 不訾诟耻。 心几烦而不绝兮, 知得王子。 山有木兮木有枝, 心悦君兮君不知。 《越人歌》是我国历史上现存的第一首译诗。 秦汉时期对“翻译官”的种种称谓: “行人”、“典客”、“大行令”、“大鸿胪”、“典乐”、“译官令”、“译官丞”等。 到汉朝,我国主要的外事活动是对北方的匈奴用兵,故翻译活动逐渐用“译”来统称了。 II.佛经翻译 佛教创立:公元前六至五世纪 创立地点:古印度 佛教流传:公元65年之前传入中国 我国的佛经翻译

中国翻译史

《英汉翻译教程》 第一章:我国翻译史简介 我国的翻译事业有约两千多年的光辉历史。早在公元前六年西汉哀帝时代,伊存到中国口传一些佛教经句,但还谈不上佛教的翻译。佛经的翻译是在东汉桓帝建和两年(公元148年)开始的,译者是安世高,他是安息人(即波斯),他译了《安般守意经》等三十多部佛经。过些时候,娄迦谶来中国,因为他是月支人,所以又称支娄迦谶。他也译了十多部佛经,但文笔生硬,不易看懂,所以从那时起,大概就有直译和意译这个问题了。他有个学生叫支亮,支亮又有个弟子叫支谦,他们三人号称“三支”,都是当时翻译佛经很有名的人。就在那时,月支派里出现了一个叫竺法护的大翻译家,他译了175部佛经,对于佛法的流传贡献很大。但这些翻译活动还只是民间私人事业。到了符秦时代,释道安设置了“译场”,成了有组织的活动,他本人不懂梵文,惟恐失真,主张严格的直译,在这期间他请来了天竺人(即印度)鸠摩罗什,他全改以前群家的直古风格,主张“意译”,他的译著为我国翻译文学奠定了基础。到南北朝时,一个叫真谛的印度佛教学者应梁武帝之聘来到中国,他译了49部经论,对中国佛教思想有较大影响。 从隋代(公元590年)起到唐代,是我国翻译事业高度发达的时期,隋代有个释彦琮,梵文造诣很深。在他以后出现了古代翻译界的巨星玄奘(与上述鸠摩罗什、真谛一起号称我国佛经三大翻译家),他成为第一个把汉文著作向国外介绍的中国人,他自创了“新译”。 从明代万历年间到清代“清学”时期,佛经翻译呈现一片衰落现象,但却出现了以徐光启、林纾(琴南)、严复(又陵)等为代表的介绍西欧各国科学、文学、哲学的翻译家。徐光启和意大利人利玛窦合作,翻译了欧几里得的《几何原本》、《测量法义》等书。林纾和他的合伙人以口述笔记的方式译了160多部文学作品,其中最著名的有《巴黎茶花女遗事》、《黑奴吁天录》、《块肉余生记》、《王子复仇记》等。严复是我国清末新兴资产阶级启蒙思想家(曾担任过北大校长),等到八国联军战役以后,他避居上海,搞翻译工作,他“曾查过汉晋六朝翻译佛经的方法”,破天荒第一次在《天演论·译例言》里正面提出了信达雅作为译事楷模。 值得一提的是,清末马建忠在他写的《拟设翻译书院议》中发挥了他所认为的“善译”的见解,可以说是试图说明翻译标准的一个开端。他的善译标准包括了三大要求:第一、译者首先要对两种语言素有研究,熟知彼此的异同;第二、弄清原文的意义精神和语气,把它传达出来;第三、译文与原文毫无出入。这些要求是很有道理的,因他本人后来没搞翻译,因此他对“善译”的见解,反被后人忽略了。 “五四”运动是我国近代翻译史的分水岭,“五四”以前最显著的表现是,以严复、林纾等为代表翻译了一系列西方资产阶级学术名著和文学作品。“五四”以后,我国翻译事业开创了一个新的历史时期,开始介绍马列主义经典著作和无产阶级文学作品,《共产党宣言》的译文就发表

中国翻译历史概况

Chapter One Introduction 中国翻译历史概况 以佛经翻译为主的古代翻译 在我国,"翻译"作为一个词出现,是在南朝的梁慧皎《高僧传》中:"先沙门法显于师子国得弥沙塞律梵本,未被翻译,而法显迁化."但是翻译的工作开始得比这个时间早得多.早在周朝时期,已经出现了专门从事翻译的官职.《周礼》中称当时的翻译官为象胥("象胥,掌蛮夷闽貉戎狄之国使,掌传王之言而喻说焉,以和亲之.")《礼记》则对负责东南西北四方的翻译人员给予了不同的称呼:"五方之民,言语不通,嗜欲不同,达其志,通其欲,东方日寄,南方日象,西方日狄,北方日译." 东汉以前,我国的翻译活动主要是各民族为沟通交流所需要的口译.佛教传人中国后,才出现了大规模的书面翻译.这种佛经的翻译肇始于东汉,发展于魏晋,到唐代臻于极盛,至宋代逐渐式微,入元代已近尾声.在这一千多年的时间里,出现了很多优秀的翻译家,如东晋时期的释道安,唐代的玄奘等.他们不但有大量的翻译实践,还提出了自己对于翻译标准和方法等方面的见解.比如释道安提出了"五失本,三不易"的翻译理论,"五失本"是指有五种情况可以允许译文不同于原文,"三不易"指翻译工作中的三种难事:难得恰当,难得契合,难得正确."五失本"与"三不易''从理论上解决了"质"与"文"的关系,即既要正确表达原著的内容和义旨,又要力求译文简洁易懂.开创佛典意译新风的是鸠摩罗什(344—413).他主张只要能存本旨,就不妨"依实出华".后人称道他的译品"善披文意,妙显径心,会达言方,风骨流便".到了唐代,佛典的翻译到了一个相当的水平.玄奘不但提出了"既须求真,又须喻俗"的翻译原则,而且还是理论和实践相结合的典范.诸如增补,省略,变位,分合,替代等一些在现代翻译教科书中常讲常练的翻译技巧,在玄奘的译经中已经运用得存乎一心,十分熟稔了. 中国近代的翻译 中国近现代翻译的第一个高潮出现在清末.1894年,《马氏文通》的作者马建忠提出"善译"的标准,要求译者对原作"所有相当之实义,委曲推究","确知其意旨之所在",而能"心悟禅解,振笔而书".简言之,译者在翻译时应做到内容与风格的高度统一.此后不久,严复在1898年出版的《天演论"译例言"》中,提出了中国历史上第一个较为明确的翻译标准,也就是"信,达,雅"的翻译标准.他的翻译活动和理论建树后来成为这一时期的翻译成就的代表. 翻译的标准 我国学者所创立的本土翻译标准 如前所说,我国历史上第一个完备的翻译标准的提出者是严复.严复,初名宗光,字又陵,号几道.1853年生于福建侯官.1866年考入福建船政学堂攻船舶驾驶专业.1871年毕业,做军舰驾驶工作.1876年被派往英国留学,1879年毕业于英国格林尼茨海军学院.同年归国并执教于母校.1880年任北洋水师学堂总教习.纳21年病逝于故里.从1898到1911年13年间是他译介生涯的鼎盛时期.严复译介的作品多系西方资产阶级著名思想家的代表作;内容涉及进化论,哲学,社会政治学,伦理学和政治经济学等领域.他的"译事三难,信,达,雅"本是他本人关于翻译的感慨,但也自然而然成了一种翻译标准.这个标准今天虽然遇到了许多不同的意见,但是仍然具有一定的指导意义." 严复之后,又有许多文学家,翻译家提出自己的关于翻译标准的观点,主要有以下几人. 伟大作家鲁迅作为一位翻译实践者,提出了自己的观点:"凡是翻译,必须兼顾着两面,一当然力求其易解,一则保存原作的风姿."(《"题未定"草》1935) 著名翻译家傅雷根据自己的文学翻译实践,在1951年提出翻译"神似说":"以效果而论,翻译应当像临画一样,所求的不在形似而在神似."(《高老头·重译本序》) 我国通学大儒钱钟书提出的翻译标准比傅雷更进一步,他的标准(毋宁说是方法)是一个"化"字.他说:"文学翻译的最高标准是'化',把作品从一国文字转变成另一国文字,既能不因语文习

中西翻译史简介

中西翻译史简介 第一节:中国翻译简史 我国翻译事业约有两千年的历史。佛经翻译始于东汉恒帝建和二年(公元148年)。迄今为止共经历了四次翻译高潮。 一、东汉至唐宋的佛经翻译:我国的佛经翻译,从东汉恒帝末年安世高译经开始,魏 晋南北朝有了进一步发展,到了唐代臻于极盛,北宋时已近式微,元以后则是尾声了。之谦的《法句经序》是我国第一篇有关翻译的论文,?最早涉及了一些重大的翻译原则?(张泽乾,1994)。道安总结了比较完善的直译原则。鸠摩罗什是主张全面意译的第一人。玄奘还提出了?既需求真,又需喻实?的翻译标准,力求忠实与通顺并举。他的?五不翻?原则总结了音译法的规律,即:(1)佛经密语须直译;(2)佛典中的多义词须音译(3)不存在相应概念的词只能音译(4)已经约定俗成的古音译保留(5)为避免语义失真用音译 二、民末清初的科技翻译:明徐光启和意大利人利马杜合作,翻译了欧几里德的《几 何原本》、《测量法义》等书。 三、鸦片战争后至“五四”前的西方政治思想和文学翻译:清林纾和他的合作者 以口述笔记的方式译了一百六十多部文学作品,其中最著名的有《黑奴吁天记》(Uncle Tom’s Cabin),《块肉余生记》(David Copperfield),《王子复仇记》(Hamlet)等(现用新译名)。严复所译作品多系西方政治经济学说,如赫胥黎的《天演论》(Evolution and Ethics and Other Essays),亚当〃斯密的《原富》(An Inquiry Into the Nature and Causes of the Wealth of Nations),斯宾塞尔的《群己权界论》(On Liberty)、甄克斯的《社会通诠》(The Study of Politics)等、并提出简洁凝练的翻译标准?信、达、雅?(faithfulness, expressiveness and elegance) 。《马氏文通》的作者马建忠于公元1894年在他的《拟设翻译书院议》中发表了他所认为的?善译?的见解,包括三大要求:第一,以者先要对两种语言素有研究,熟知彼此的一统;第二,龙庆原文的意义、精神和语气,把它传达出来;第三,译文与原文毫无出入,?译成之文,适如其所译?。 四、民国时期的翻译事业的繁荣:鲁迅先生认为,?凡是翻译,必须兼顾两个方面, 一则当然其义易解,一则保存原作的丰姿。?并提出了通行的翻译标准:忠实于通顺(faithfulness and smoothness)。瞿秋白论证翻译是可以做到又信又顺的。林语堂提出了?忠实的标准,通顺的标准,美的标准?。傅雷的?重形似而不重神似?的标准。钱钟书提出的?精神姿致依然故我?和?化境?之说。哲学家艾思奇则总结说,?翻译的原则总不外是以‘信’为最根本的东西,‘达’和‘雅’的对于‘信’,就像属性对于本质一样,是分不开的然而是第二义的存在。 五、第五次翻译高潮:无论在翻译的规模和译作的数量上都远远超过了以前任何时期。第二节西方翻译简史:古代翻译活动、近代翻译活动、当代翻译活动 一、古代翻译活动: 西方翻译活动可追溯到公元前三世纪。当时有文字记载的翻译作品已经问世:七十二位犹太学者在自己埃及的亚历山大城翻译了《圣经〃旧约》,即后人所称的《七十子希腊文本》;

张培基《英汉翻译教程》(修订本)配套题库(我国翻译史简 介)

第1章我国翻译史简介 一、我国翻译事业的历史有多久?对我国的翻译史进行大致划分,并给出具有代表性的翻译家。 【答案】 中国的翻译,从公元67年天竺僧侣(摄摩腾和和竺法兰)到洛阳白马寺讲经以来,已有近两千年的历史,出现过的翻译高潮大致有五次。 第一次:从东汉到唐宋的佛经翻译。这时期出现过中国佛经的三大翻译家:鸠摩罗什、真谛、玄奘。 第二次:明末清初的西方科技著作的汉译和中国典籍的西译。翻译家有徐光启、利玛窦、汤若望、南怀仁等。 第三次:五四以前对西方政治、哲学和文学作品的翻译。翻译家有林纾、严复、梁启超等。 第四次:1949年中华人民共和国建国初后的十几年。这个时期我国对马列著作的汉译和《毛泽东选集》民族经文及外文的翻译投入了大量的人力。翻译家有鲁迅、赵元任、朱生豪、林语堂等。 第五次:1978年实行改革开放政策后开始的西方学术著作和文艺作品的大量翻译,是中国翻译史迄今为止的第五次高潮。翻译家有傅雷、钱钟书、杨绛等。 【解析】我国翻译史、所出现五次高潮、每一个高潮的研究内容、代表译家是每一个译者都应了解的基本内容,此外译者也应对西方翻译历史有一个大致的了解,以便更好地开展翻译

实践。 二、简要介绍严复“信、达、雅”的翻译标准并谈谈你对这个标准的看法。 【答案】 严复在《天演论》译文的例言中提出了“信、达、雅”的翻译原则和标准。“信”是指对原文和译文两个关联事物的可靠性和一致性,即译文要忠实于原文的内容、风格、思想以及精神,不可歪曲原文,不可遗漏原文的重要内容。“达”是指译文要通达流畅,符合现代汉语表达习惯,符合汉语的语法规范。“雅”是指译文所要达到的文学美感,即文采气质。译文语句要规范、得体、生动、优美,有独特的文学典雅气质。 我认为作为一名合格的译员,首先应该保证前两个标准,即“信”“达”的实现。在翻译时,应确保译文完整准确地传达原文的意思、符合译入语的语法规范,然后在有余力的情况下,再去完善译文,使之更具美感。 三、新中国成立后的翻译工作有哪些特点? 【答案】 ①翻译工作者在党的领导下,有组织、有计划、有系统地进行工作,逐渐取代了抢译、乱译和重复浪费的现象; ②翻译作品质量大大提高,逐渐克服了粗枝大叶、不负责任的风气; ③翻译工作者为了更好地为社会主义建设服务,开展了批评与自我批评,逐渐消除了过去各种不良现象和无人过问的状况; ④翻译工作者不仅肩负着外译汉的任务,同时为了宣传马列主义、毛泽东思想,介绍我国社会主义革命和建设的经验以及我国优秀的文化遗产,还肩负了汉译外的任务;

中国翻译史笔记

中国翻译史 袁素平 一.世界四大文明古国 中国,印度,巴比伦,埃及 中国文化的特色:从未中断过。 巴比伦灭了 印度,埃及曾经沦为殖民地 中华文化生生不息 季羡林:“中华文化之所以能保持永保青春,万灵之药就是翻译。” 中华文化的源头: 易经the art of changes Confucius 孔子 Mencius 孟子 Confucianism 儒家文化 太极生两仪include 一生二,二生三,三生万物 西方之水: 第一次高峰:从明朝末年开始引进西方的文化,最终在清朝末年特别是在于1840年鸦片战

争之后达到高潮。清朝末年其实是引起了中华文化的后退。 梁启超:中国文化的集大成者。掀开了翻译政治小说之幕,林纾翻译文艺小说。 第二次高峰:以1919年为界,掀起第二次高峰 鲁迅是现代小说之父,其一半的活动是和翻译相关的。 第三次高峰:改革开放1978年之后。近现代出不了大家。 中西文化之别: 中国重智即动头脑的能力 外国重用即动手实践能力 因此要:手脑联盟 四次翻译高潮: 1.东汉至唐宋时期,佛经翻译盛行。、 2.明末清初,欧洲一批耶稣会士相继来华进行翻译活动,主要以传教为宗旨,同时也介绍了西方学术。 3.鸦片战争至“五四运动”期间的西方思想和文学翻译。这一时期最引人瞩目的就是严复和林纾。(3,4之间是断档的历史)4.改革开放至今 东汉至唐宋时期的佛经翻译 1.东汉桓帝年间的安世高:《安般守意经》等三十五部佛经,开

后世禅学之源,其译本“义理明晰,文字允正,辩而不华,质 而不野”但其主要偏于直译。 2.支娄迦谶:其所译经典,译文流畅,但为了力求促使原来面目,“辞质多胡音”,即多音译。 3.支谦:“颇从文丽”,开创了不踏实原著的译风,对三国至西晋的翻译产生了很大的影响。翻译中的“会译”体裁,以及用音 译取代前期的音译。(复译,重译等) 4.中国佛经翻译史上,由此一直存在“质朴”和“文丽”两派。北魏孝文帝:促进中华文化的融合 千古一后:冯太后 前秦:释道安 前秦时代:佛经翻译由私人事业转入了译场翻译,释道安在朝廷的支持下首创译场翻译制度,采用“会译”法来研究翻译。他主张严格的直译,并总结汉末以来的译经经验,提出了著名的“五失本,三不易”理论,指出五种容易使译文推动原来面目的情况和三种不容易处理的情况。 译胡为秦,有五失本也: 一者,胡语尽倒(词序颠倒),而使从秦(改从汉语语序),一失本也。二者,胡经尚质,秦人好文(采),传可众心,非文不合,斯二失本

中国翻译史第二个时期

中国翻译史第二个时期 这个时期不论在翻译内容,还是翻译方法广都较第一个时期有很大变化。 这个时期又可分为三个阶段: 1.明朗万历初年,意大利人利玛窦来小国开始介绍西洋科学知识,从事科学者作 翻译到明末清初的翻译。 2清未同文馆及江南制造局的翻译。 3严复、林纤等人开始两洋社会政治作品及文学作品的翻译。 明万历几年(公入1581年)意大利传教土利玛窦来到小国,先后公南吕、南京、北京等地建教堂传教,学汉话语吉、文学和礼仪,主张将孔孟之道与敬祖同天主教义相融合,并介绍西方文化。利玛窦深通汉语,曾译过四书寄回意大利,对促进巾西方文化 交流起厂重要作用c明末清初中冈出现了两位有影响的科学家与翻译家徐光肩(1562—1633)和李之藻(1565—1630)。徐光肩,字子先,号玄层,上海哨人.万历年间为进 士c李之藻,字振之,杭州人,万历年为进士。他们生活在明朝末年,当时朱家于朝 昏庸腐败、内外交闲,而多数士大夫们只是醉心仕途,空谈理学。徐、李虽身入宦海 而不染恶习,孜放不倦研究农学.后从利玛安学习天文、历法、数学等dy方科学知识。徐光启曾与利玛窦合译《几何原本》、仰Ij量法义》等。李之藻的译著极丰.先后与 利玛窦翻译两方著作20多种、54卷,共中“器编”10种全部介绍西方科学知识。 清朝统治中国以后,对外采取闭关白守,对内奉行愚民政策。所以在鸦片战争前 200多年中,几乎无翻译事、Ik可言,严重地禁铜了中四方学术文化的交流。 鸦片战争以后,中同沦为半封建半殖民地国家。清政府为广府酬外交的需要,不得 不同外国人打交道。因此,于1863年在北京,次年在上海、广州没穴了问义馆,从事翻泽丁作。深圳翻译公司 北京的向文馆先设立英文馆,后增设法文馆,俄罗斯文馆.德文馆及科学馆。各馆 毕业生除留馆工作外.大部分充仟口译或笔译工作G所以,同文馆一入面是清政府培 养翻译人才的学校,另一方圆又是翻译机关。后来同文馆归于京师大学堂,光绪二卜 亡年改名为译学馆,它的第一位主持人就是严复。 继向文馆之后,于1867年在上海设立了江南制造局。1869年在局内义成立了翻译局,从事以F1然科学为主要内奔的翻泽活动。 晚清和民国初年,社会政治作品及文学作品也随着人们对于社会改革需求的增长而 逐渐发展起来。

中西翻译史简述

中西翻译史简述* 1.History of Translation in China ?1)佛经翻译时期,从东汉开始至唐宋,1000余年(三大佛经翻译家:鸠摩罗什,真谛,玄奘) ?2)明朝后期:西方科技翻译 ?3)清朝后期:西方哲学、文学翻译 重要人物:林纾、严复 ?4)中国近代翻译史:“五四”运动是分水岭; 重要人物:鲁迅、瞿秋白 ?5)繁荣阶段:新中国的成立 ?The Four Major Waves of Translation in China The Translation of Buddhist Scriptures The Beginning Stage The Development Stage The Climax Stage The Concluding Stage The Translation of Western Science and Technology since 16th Century The Translation of Western Philosophy and Literature since 19th Century The Translation since May 4th Movement 翻译在我国有悠久的历史。有史记载的翻译活动也与宗教直接有关。佛教早在上古时期便在亚洲诸国传播,东汉时传入我国。以后近千年内,我国共翻译佛经1500多部,丰富的翻译实践造就了一批佛经翻译理论家,他们是三国时期(公元3世纪)的支谦、东晋的道安、六朝时代的鸠摩罗什,隋代的彦综和唐代的玄奘。支谦的《法句经序》是我国第一篇有关翻译的论文,道安总结了比较完善的直译原则,鸠摩罗什是主张全面意译的第一人。彦综和玄奘生活在佛经翻译的鼎盛时期,玄奘亲自去印度取经,带回经书600余卷,其中一半以上由他+译出,译文具有“意译直译、圆满调和”的品味。他还提出了“既须求真,又须喻俗”的翻译标准,力求忠实与易懂并重。 东汉至唐宋的佛经翻译是五四运动前的第一次翻译高潮。第二次翻译高潮是明末清初的

中国翻译史简介

中国翻译史简介 我国的翻译史可上溯到西汉哀帝时代,差不多有两千年的历史。我国翻译事业的发展大致经历了以下几个阶段: 1.1第一阶段: 东汉桓(huan)帝建和二年(公元一四八年)到南北朝,历经四百多年。中国翻译史的一页是从翻译佛经时揭开的,译者是安世高,波斯(今伊朗)人,他译了《安般守意经》等三十多部佛经。后来月支国(西域)人娄迦谶(chen)来中国,他译了十多部佛经。支娄迦谶译笔生硬,读者不易看懂,所以从那时起,大概就有直译和意译这一类问题了。他有个学生叫支亮,支亮有个弟子叫支谦,他们三人号称“三支”,都是当时翻译佛经很有名的人。就在那时,月支派里还出现了一个名叫竺法护的大翻译家,他译了一百七十五部佛经,对于佛法的流传贡献很大。这些翻译活动还只是民间行为。到了符秦时代,翻译事业出现了一大进步,成为有组织的活动。在释道安的主持下设置了译场。道安自己不懂梵文,惟恐翻译失真,主张严格的直译,因此在他主持下翻译的《鞞(bi)婆沙》便是一字一句地翻译下来的。道安在这期间请来了天竺(即印度)人鸠摩罗什。鸠摩罗什考证了以前的佛经译者,批评了翻译的文体,检讨了翻译的方法。他全改以前群家的古直风格,主张意译,改正了过去音译(transliteration)的弱点,并提倡译者署名。再后来,一个名叫真谛的印度佛教学者来到中国,译了四十九部经论,对中国佛教思想有较大影响。 1.2第二阶段: 从随代(公元五九O年)起到唐代,是我国翻译事业的鼎盛时期。隋代历史较短,译者和译经都不多,其中有释彦琮(cong)(俗姓李,赵郡人)者,梵文造诣很深。他对于翻译理论,曾有比较透彻的发挥,认为译者应该:(一)“诚心爱法,志愿益人,不惮久时”;(二)“襟(jin)抱平恕,器量虚融,不好专执”,(三)“耽于道术,澹(dan)于名利,不欲高炫”。在他以后,出现了古代翻译界的巨星玄奘(与上述鸠摩罗什、真谛一起号称我国佛教三大翻译家)。玄奘在唐太宗贞观二年(公元六二八年)出发去印度(天竺zhu)求经,十七年后才回国。他带回佛经六百五十七部,主持了比过去在组织制度方面更为健全的译场,在十九年间译出七十五部佛经,共一三三五卷。他不但把佛经由梵文译成汉文,而且把老子著作的一部分译成梵文,成为第一个把汉文著作向国外介绍的中国人。玄奘在翻译理论方面也是有贡献的,他所提出的翻译标准“既须求真,又须喻俗”,意即“忠实、通顺”,直到今天仍然有指导意义。 1.3第三阶段: 从明代万历年间到清代“新学”时期,在佛经翻译呈现一片衰落现象的同时,却出现了以徐光启、林纾(shu)(琴南)、严复(又陵)等为代表的介绍西欧各国科学、文学、哲学的翻译家。严复是我国清末新兴资产阶级的启蒙思想家。他从光绪二十四年到宣统三年(公元一八九八年到一九—一年)这十三年间潜心翻译,所译作品多系西方政治经济学说。关于严复的翻译方法,鲁迅曾在《二心集》里说过,严复“曾经查过汉晋六朝翻译佛经的方法”。严复参照古代翻译佛经的经验,根据自己翻译的实践,在《天演论》(公元一八九八年出版)卷首的《译例言》中提出了著名的“信、达、雅”翻译标准。严复曾说:“译事三难:信、达、雅(faithfulness, expressiveness and elegance)求其信,已大难矣!顾信矣,不达,虽译,犹不译也,则达尚焉。”他主张的“信”是“意义不倍(背)本文”,“达”是不拘泥于原文形式,尽译文语言的能事以求原意明显,为“达”也是为“信”,两者是统一的。但严复对“雅”字的解释今天看来是不足取的。他所谓的“雅”,是指脱离原文而片面追求译文本身的古雅。他认为只有译文本身采用上等的文言文,才算登大雅之堂。严复自己在翻译实践中所遵循的也是“与其伤雅,毋宁失真”,因而译文不但艰深难懂,又不忠实于原文,类似改编。由于时代不同,严复对“信、达、雅”翻译标准的解释有一定的局限。作为翻译标准,

相关文档