diff --git a/ast/ast.go b/ast/ast.go
index 4567d33..06c0531 100644
--- a/ast/ast.go
+++ b/ast/ast.go
@@ -310,3 +310,25 @@ func (ie *IndexExpression) String() string {
 
 	return out.String()
 }
+
+type HashLiteral struct {
+	Token token.Token // The '{' token
+	Pairs map[Expression]Expression
+}
+
+func (hl *HashLiteral) expressionNode()      {}
+func (hl *HashLiteral) TokenLiteral() string { return hl.Token.Literal }
+func (hl *HashLiteral) String() string {
+	var out bytes.Buffer
+
+	var pairs []string
+	for key, value := range hl.Pairs {
+		pairs = append(pairs, key.String()+token.Colon+value.String())
+	}
+
+	out.WriteString(token.LeftBrace)
+	out.WriteString(strings.Join(pairs, token.Comma+" "))
+	out.WriteString(token.RightBrace)
+
+	return out.String()
+}
diff --git a/evaluator/evaluator.go b/evaluator/evaluator.go
index 4eee2fd..5a3b2a9 100644
--- a/evaluator/evaluator.go
+++ b/evaluator/evaluator.go
@@ -100,6 +100,8 @@ func Eval(node ast.Node, env *object.Environment) object.Object {
 			return elements[0]
 		}
 		return &object.Array{Elements: elements}
+	case *ast.HashLiteral:
+		return evalHashLiteral(node, env)
 	}
 
 	return nil
@@ -259,6 +261,8 @@ func evalIndexExpression(left, index object.Object) object.Object {
 	switch {
 	case left.Type() == object.ArrayObj && index.Type() == object.IntegerObj:
 		return evalArrayIndexExpression(left, index)
+	case left.Type() == object.HashObj:
+		return evalHashIndexExpression(left, index)
 	default:
 		return newError("index operator not supported: %s", left.Type())
 	}
@@ -276,6 +280,22 @@ func evalArrayIndexExpression(array, index object.Object) object.Object {
 	return arrayObject.Elements[idx]
 }
 
+func evalHashIndexExpression(hash, index object.Object) object.Object {
+	hashObject := hash.(*object.Hash)
+
+	key, ok := index.(object.Hashable)
+	if !ok {
+		return newError("unusable as hash key: %s", index.Type())
+	}
+
+	pair, ok := hashObject.Pairs[key.HashKey()]
+	if !ok {
+		return Null
+	}
+
+	return pair.Value
+}
+
 func evalIdentifier(node *ast.Identifier, env *object.Environment) object.Object {
 	if val, ok := env.Get(node.Value); ok {
 		return val
@@ -288,6 +308,32 @@ func evalIdentifier(node *ast.Identifier, env *object.Environment) object.Object
 	return newError("%s: %s", identifierNotFoundError, node.Value)
 }
 
+func evalHashLiteral(node *ast.HashLiteral, env *object.Environment) object.Object {
+	pairs := make(map[object.HashKey]object.HashPair)
+
+	for keyNode, valueNode := range node.Pairs {
+		key := Eval(keyNode, env)
+		if isError(key) {
+			return key
+		}
+
+		hashKey, ok := key.(object.Hashable)
+		if !ok {
+			return newError("unusable as hash key: %s", key.Type())
+		}
+
+		value := Eval(valueNode, env)
+		if isError(value) {
+			return value
+		}
+
+		hashed := hashKey.HashKey()
+		pairs[hashed] = object.HashPair{Key: key, Value: value}
+	}
+
+	return &object.Hash{Pairs: pairs}
+}
+
 func applyFunction(fn object.Object, args []object.Object) object.Object {
 	switch fn := fn.(type) {
 	case *object.Function:
diff --git a/evaluator/evaluator_test.go b/evaluator/evaluator_test.go
index 8a1adb4..33bd043 100644
--- a/evaluator/evaluator_test.go
+++ b/evaluator/evaluator_test.go
@@ -159,6 +159,7 @@ func TestErrorHandling(t *testing.T) {
 		`, fmt.Sprintf("%s: %s + %s", unknownOperatorError, object.BooleanObj, object.BooleanObj)},
 		{"foobar", fmt.Sprintf("%s: %s", identifierNotFoundError, "foobar")},
 		{`"Hello" - "World"`, fmt.Sprintf("%s: %s - %s", unknownOperatorError, object.StringObj, object.StringObj)},
+		{`{"name": "Monky"}[fn(x) { x }];`, fmt.Sprintf("unusable as hash key: %s", object.FunctionObj)},
 	}
 
 	for _, tt := range tests {
@@ -401,6 +402,94 @@ func TestArrayIndexExpressions(t *testing.T) {
 	}
 }
 
+func TestHashLiterals(t *testing.T) {
+	input := `
+		let two = "two";
+		{
+			"one": 10 - 9,
+			"two": 1 + 1,
+			"thr" + "ee": 6 / 2,
+			4: 4,
+			true: 5,
+			false: 6,
+		}
+	`
+
+	evaluated := testEval(input)
+	result, ok := evaluated.(*object.Hash)
+	if !ok {
+		t.Fatalf("Eval didn't return %T. got=%T (%+v)", object.Hash{}, evaluated, evaluated)
+	}
+
+	expected := map[object.HashKey]int64{
+		(&object.String{Value: "one"}).HashKey():   1,
+		(&object.String{Value: "two"}).HashKey():   2,
+		(&object.String{Value: "three"}).HashKey(): 3,
+		(&object.Integer{Value: 4}).HashKey():      4,
+		True.HashKey():                             5,
+		False.HashKey():                            6,
+	}
+
+	if len(result.Pairs) != len(expected) {
+		t.Fatalf("wrong num of pairs. got=%d", len(result.Pairs))
+	}
+
+	for expectedKey, expectedValue := range expected {
+		pair, ok := result.Pairs[expectedKey]
+		if !ok {
+			t.Errorf("no pair for given key in pairs")
+		}
+
+		testIntegerObject(t, pair.Value, expectedValue)
+	}
+}
+
+func TestHashIndexExpressions(t *testing.T) {
+	tests := []struct {
+		input    string
+		expected interface{}
+	}{
+		{
+			`{"foo": 5}["foo"]`,
+			5,
+		},
+		{
+			`{"foo": 5}["bar"]`,
+			nil,
+		},
+		{
+			`let key = "foo"; {"foo": 5}[key]`,
+			5,
+		},
+		{
+			`{}["foo"]`,
+			nil,
+		},
+		{
+			`{5: 5}[5]`,
+			5,
+		},
+		{
+			`{true: 5}[true]`,
+			5,
+		},
+		{
+			`{false: 5}[false]`,
+			5,
+		},
+	}
+
+	for _, tt := range tests {
+		evaluated := testEval(tt.input)
+		integer, ok := tt.expected.(int)
+		if ok {
+			testIntegerObject(t, evaluated, int64(integer))
+		} else {
+			testNullObject(t, evaluated)
+		}
+	}
+}
+
 func testEval(input string) object.Object {
 	l := lexer.New(input)
 	p := parser.New(l)
diff --git a/lexer/lexer.go b/lexer/lexer.go
index 0e57684..a3eb78b 100644
--- a/lexer/lexer.go
+++ b/lexer/lexer.go
@@ -57,6 +57,8 @@ func (l *Lexer) NextToken() token.Token {
 		tok = newToken(token.Comma, l.ch)
 	case ';':
 		tok = newToken(token.Semicolon, l.ch)
+	case ':':
+		tok = newToken(token.Colon, l.ch)
 	case '(':
 		tok = newToken(token.LeftParen, l.ch)
 	case ')':
diff --git a/lexer/lexer_test.go b/lexer/lexer_test.go
index 26423c4..943062c 100644
--- a/lexer/lexer_test.go
+++ b/lexer/lexer_test.go
@@ -29,6 +29,7 @@ func TestNextToken(t *testing.T) {
 		"foobar"
 		"foo bar"
 		[1, 2];
+		{"foo": "bar"}
 	`
 
 	tests := []struct {
@@ -116,6 +117,11 @@ func TestNextToken(t *testing.T) {
 		{token.Int, "2"},
 		{token.RightBracket, "]"},
 		{token.Semicolon, ";"},
+		{token.LeftBrace, "{"},
+		{token.String, "foo"},
+		{token.Colon, ":"},
+		{token.String, "bar"},
+		{token.RightBrace, "}"},
 		{token.EOF, ""},
 	}
 
diff --git a/object/object.go b/object/object.go
index d997024..a6fd047 100644
--- a/object/object.go
+++ b/object/object.go
@@ -3,9 +3,11 @@ package object
 import (
 	"bytes"
 	"fmt"
+	"hash/fnv"
 	"strings"
 
 	"github.com/kitasuke/monkey-go/ast"
+	"github.com/kitasuke/monkey-go/token"
 )
 
 type ObjectType string
@@ -20,8 +22,18 @@ const (
 	StringObj      = "String"
 	BuiltinObj     = "Builtin"
 	ArrayObj       = "Array"
+	HashObj        = "Hash"
 )
 
+type HashKey struct {
+	Type  ObjectType
+	Value uint64
+}
+
+type Hashable interface {
+	HashKey() HashKey
+}
+
 type Object interface {
 	Type() ObjectType
 	Inspect() string
@@ -33,6 +45,9 @@ type Integer struct {
 
 func (i *Integer) Type() ObjectType { return IntegerObj }
 func (i *Integer) Inspect() string  { return fmt.Sprintf("%d", i.Value) }
+func (i *Integer) HashKey() HashKey {
+	return HashKey{Type: i.Type(), Value: uint64(i.Value)}
+}
 
 type Boolean struct {
 	Value bool
@@ -40,6 +55,17 @@ type Boolean struct {
 
 func (b *Boolean) Type() ObjectType { return BooleanObj }
 func (b *Boolean) Inspect() string  { return fmt.Sprintf("%t", b.Value) }
+func (b *Boolean) HashKey() HashKey {
+	var value uint64
+
+	if b.Value {
+		value = 1
+	} else {
+		value = 0
+	}
+
+	return HashKey{Type: b.Type(), Value: value}
+}
 
 type Null struct{}
 
@@ -91,6 +117,12 @@ type String struct {
 
 func (s *String) Type() ObjectType { return StringObj }
 func (s *String) Inspect() string  { return s.Value }
+func (s *String) HashKey() HashKey {
+	h := fnv.New64a()
+	h.Write([]byte(s.Value))
+
+	return HashKey{Type: s.Type(), Value: h.Sum64()}
+}
 
 type BuiltinFunction func(args ...Object) Object
 
@@ -120,3 +152,28 @@ func (ao *Array) Inspect() string {
 
 	return out.String()
 }
+
+type HashPair struct {
+	Key   Object
+	Value Object
+}
+
+type Hash struct {
+	Pairs map[HashKey]HashPair
+}
+
+func (h *Hash) Type() ObjectType { return HashObj }
+func (h *Hash) Inspect() string {
+	var out bytes.Buffer
+
+	var pairs []string
+	for _, pair := range h.Pairs {
+		pairs = append(pairs, fmt.Sprintf("%s: %s", pair.Key.Inspect(), pair.Value.Inspect()))
+	}
+
+	out.WriteString(token.LeftBrace)
+	out.WriteString(strings.Join(pairs, token.Comma+" "))
+	out.WriteString(token.RightBrace)
+
+	return out.String()
+}
diff --git a/object/object_test.go b/object/object_test.go
new file mode 100644
index 0000000..ce530b0
--- /dev/null
+++ b/object/object_test.go
@@ -0,0 +1,16 @@
+package object
+
+import "testing"
+
+func TestStringHashKey(t *testing.T) {
+	hello1 := &String{Value: "Hello World"}
+	hello2 := &String{Value: "Hello World"}
+	diff1 := &String{Value: "My name is johnny"}
+	diff2 := &String{Value: "My name is johnny"}
+
+	if hello1.HashKey() != hello2.HashKey() ||
+		diff1.HashKey() != diff2.HashKey() ||
+		hello1.HashKey() == diff1.HashKey() {
+		t.Errorf("strings with same content have different hash keys")
+	}
+}
diff --git a/parser/parser.go b/parser/parser.go
index f2f182b..8825aab 100644
--- a/parser/parser.go
+++ b/parser/parser.go
@@ -65,6 +65,7 @@ func New(l *lexer.Lexer) *Parser {
 	p.registerPrefix(token.Function, p.parseFunctionLiteral)
 	p.registerPrefix(token.String, p.parseStringLiteral)
 	p.registerPrefix(token.LeftBracket, p.parseArrayLiteral)
+	p.registerPrefix(token.LeftBrace, p.parseHashLiteral)
 
 	p.infixParseFns = make(map[token.TokenType]infixParseFn)
 	p.registerInfix(token.Plus, p.parseInfixExpression)
@@ -386,6 +387,35 @@ func (p *Parser) parseIndexExpression(left ast.Expression) ast.Expression {
 	return exp
 }
 
+func (p *Parser) parseHashLiteral() ast.Expression {
+	hash := &ast.HashLiteral{Token: p.currentToken}
+	hash.Pairs = make(map[ast.Expression]ast.Expression)
+
+	for !p.peekTokenIs(token.RightBrace) {
+		p.nextToken()
+		key := p.parseExpression(Lowest)
+
+		if !p.expectPeek(token.Colon) {
+			return nil
+		}
+
+		p.nextToken()
+		value := p.parseExpression(Lowest)
+
+		hash.Pairs[key] = value
+
+		if !p.peekTokenIs(token.RightBrace) && !p.expectPeek(token.Comma) {
+			return nil
+		}
+	}
+
+	if !p.expectPeek(token.RightBrace) {
+		return nil
+	}
+
+	return hash
+}
+
 // Infix expressions
 
 func (p *Parser) parseInfixExpression(left ast.Expression) ast.Expression {
diff --git a/parser/parser_test.go b/parser/parser_test.go
index 77e5ab8..9dbaff7 100644
--- a/parser/parser_test.go
+++ b/parser/parser_test.go
@@ -523,6 +523,108 @@ func TestStringLiteralExpression(t *testing.T) {
 	}
 }
 
+func TestParsingHashLiteralsStringKeys(t *testing.T) {
+	input := `{"one": 1, "two": 2, "three": 3}`
+
+	l := lexer.New(input)
+	p := New(l)
+	program := p.ParseProgram()
+	checkParserErrors(t, p)
+
+	stmt := program.Statements[0].(*ast.ExpressionStatement)
+	hash, ok := stmt.Expression.(*ast.HashLiteral)
+	if !ok {
+		t.Fatalf("exp is not %T. got=%T", ast.HashLiteral{}, stmt.Expression)
+	}
+
+	if len(hash.Pairs) != 3 {
+		t.Errorf("hash.Pairs has wrong length. got=%d", len(hash.Pairs))
+	}
+
+	expected := map[string]int64{
+		"one":   1,
+		"two":   2,
+		"three": 3,
+	}
+
+	for key, value := range hash.Pairs {
+		literal, ok := key.(*ast.StringLiteral)
+		if !ok {
+			t.Errorf("key is not %T. got=%T", ast.StringLiteral{}, key)
+		}
+
+		expectedValue := expected[literal.String()]
+
+		testIntegerLiteral(t, value, expectedValue)
+	}
+}
+
+func TestParsingEmptyHashLiteral(t *testing.T) {
+	input := `{}`
+
+	l := lexer.New(input)
+	p := New(l)
+	program := p.ParseProgram()
+	checkParserErrors(t, p)
+
+	stmt := program.Statements[0].(*ast.ExpressionStatement)
+	hash, ok := stmt.Expression.(*ast.HashLiteral)
+	if !ok {
+		t.Fatalf("exp is not %T. got=%T", ast.HashLiteral{}, stmt.Expression)
+	}
+
+	if len(hash.Pairs) != 0 {
+		t.Errorf("hash.Pairs has wrong length. got=%d", len(hash.Pairs))
+	}
+}
+
+func TestParsingHashLiteralsWithExpressions(t *testing.T) {
+	input := `{"one": 0 + 1, "two": 10 - 8, "three": 15 / 5}`
+
+	l := lexer.New(input)
+	p := New(l)
+	program := p.ParseProgram()
+	checkParserErrors(t, p)
+
+	stmt := program.Statements[0].(*ast.ExpressionStatement)
+	hash, ok := stmt.Expression.(*ast.HashLiteral)
+	if !ok {
+		t.Fatalf("exp is not %T. got=%T", ast.HashLiteral{}, stmt.Expression)
+	}
+
+	if len(hash.Pairs) != 3 {
+		t.Errorf("hash.Pairs has wrong length. got=%d", len(hash.Pairs))
+	}
+
+	tests := map[string]func(ast.Expression){
+		"one": func(e ast.Expression) {
+			testInfixExpression(t, e, 0, "+", 1)
+		},
+		"two": func(e ast.Expression) {
+			testInfixExpression(t, e, 10, "-", 8)
+		},
+		"three": func(e ast.Expression) {
+			testInfixExpression(t, e, 15, "/", 5)
+		},
+	}
+
+	for key, value := range hash.Pairs {
+		literal, ok := key.(*ast.StringLiteral)
+		if !ok {
+			t.Errorf("key is not %T. got=%T", ast.StringLiteral{}, key)
+			continue
+		}
+
+		testFunc, ok := tests[literal.String()]
+		if !ok {
+			t.Errorf("No test function for key %q found", literal.String())
+			continue
+		}
+
+		testFunc(value)
+	}
+}
+
 func createParseProgram(input string, t *testing.T) *ast.Program {
 	l := lexer.New(input)
 	p := New(l)
diff --git a/token/token.go b/token/token.go
index d70234b..ad6c8f3 100644
--- a/token/token.go
+++ b/token/token.go
@@ -32,6 +32,7 @@ const (
 	// Delimiters
 	Comma     = ","
 	Semicolon = ";"
+	Colon     = ":"
 
 	LeftParen    = "("
 	RightParen   = ")"